Potencial del acoplamiento GC-MS para la determinación

Universidad Jaume I Departamento de Química Física y Analítica

Instituto Universitario de Plaguicidas y Aguas

POTENCIAL DEL ACOPLAMIENTO GC-MS

PARA LA DETERMINACIÓN DE CONTAMINANTES ORGÁNICOS

EN MUESTRAS DE MATRIZ COMPLEJA

Tesis Doctoral

JAIME NÁCHER MESTRE

2011

Dr. Roque Serrano Gallego, Profesor Titular de Química Analítica de la Universitat Jaume I

de Castellón y Dr. Jaume Pérez Sánchez, Profesor de Investigación del Instituto de

Acuicultura de Torre la Sal del Consejo Superior de Investigaciones Científicas,

Certifican: que la Tesis Doctoral “Potencial del acoplamiento GC-MS para la determinación

de contaminantes orgánicos en muestras de matriz compleja” ha sido desarrollada bajo su

dirección, en el Instituto Universitario de Plaguicidas y Aguas, Departamento de Química

Física y Analítica de la Universitad Jaume I de Castellón, por Jaime Nácher Mestre.

Lo que certificamos para los efectos oportunos en Castellón de la Plana, a 11 de Enero de

2011.

Fdo. Dr. Roque Serrano Gallego Fdo. Dr. Jaume Pérez Sánchez

Esta Tesis ha sido realizada, y consecuentemente será defendida, para la obtención del

título de Doctorado en Química Analítica de la Universidad Jaume I.

Previamente a la defensa de la Tesis Doctoral, este trabajo ha sido evaluado por tres

censores independientes directamente relacionados con el área de investigación, Dra.

Maria Teresa Galceran Huguet (Catedrática del Departamento de Química Analítica de la

Univesidad de Barcelona), Dra. Rosa Maria Marcé Recasens (Catedrática del Departamento

de Química Analítica de la Univesitad Rovira i Virgili de Tarragona) y Dra. Inmaculada Varó

Vaello (Científica Titular del Instituto de Acuicultura de Torre de la Sal del Consejo Superior

de Investigaciones Científicas).

Agradecimientos

Desde mi comienzo en el mundo de la investigación a finales del 2006 hasta hoy, han

sido muchas las personas que me han acompañado, un grupo fuerte que me ha respaldado

en todo momento. Familiares, tutores, compañeros y amigos, todos en conjunto han

contribuido de alguna manera a este propósito y por ello les estoy enormemente

agradecido.

En primer lugar, querría agradecer a mis directores de Tesis. Al Dr. Roque Serrano por

confiar en mí para el proyecto de investigación y darme la oportunidad de realizar la Tesis

Doctoral en el IUPA. También su dirección y dedicación continua han sido determinantes en

mi carrera investigadora. Al Dr. Jaume Pérez por su dedicación y haber confiado en mi

trabajo. Gracias también por tus aportaciones científicas desde el mundo de la acuicultura.

Querría agradecer al Dr. Félix Hernández su dedicación constante y colaboración

recibida así como todas las oportunidades que he tenido desde mi comienzo en la

investigación.

A Sandra, por tenerla siempre a mi lado, ser de gran apoyo y buena consejera en

todo momento. También a mis padres Marimer y Pedro y a los cracks de mis hermanos

Pedro y Diego, pues sin los cinco hubiera sido imposible este camino.

Los compañeros de investigación han sido fundamentales. El continuo apoyo y ayuda

recibida durante todos estos años me ha ayudado a trabajar en un clima de trabajo muy

saludable. En este sentido querría agradecer al gran número de compañeros y amigos con

los que he compartido horas en el IUPA. Me acuerdo de Ana María, Ángel, Arantxa, Carlos,

Carmen, Cecilia, Clara, Cristina, Edu, Elena P., Elena S., Emma, Federica, Inés, Jose, Juanvi,

Laura, Luna, Maria, Mari Carmen, Mercedes, Miguel Angel, Mónica, Neus, Paco, Ramón,

Robert, Sandra, Sergi, Susana, Tania, Tatiana, Toni, Vima y Ximo.

También querría agradecer la colaboración y apoyo del grupo del IATS, entre ellos a

Alfonso, Azucena, Josep, Juan Carlos, Laura, Mª Angeles y Paco Amat.

A todos, muchas gracias. He aprendido muchísimo.

Resumen

A lo largo de esta Tesis Doctoral se investiga el potencial analítico y la aplicabilidad

del acoplamiento instrumental GC-MS mediante diferentes analizadores de masa.

La Tesis Doctoral consta de tres partes diferenciadas. En la primera se desarrolla,

valida y aplica metodología analítica basada en el acoplamiento GC-MS para la

determinación de contaminantes orgánicos persistentes en muestras procedentes de la

acuicultura marina. Entre las muestras se incluyen materias primas de origen vegetal y

animal, piensos con diferente composición y filetes de dorada cultivada. Concretamente se

desarrollan metodologías para la determinación de PAHs y PBDEs. Para ello se utilizaran

analizadores de triple cuadrupolo y tiempo de vuelo. La utilización de un analizador de

triple cuadrupolo como herramienta de análisis ofreció una elevada sensibilidad y

selectividad para el análisis de compuestos target a niveles traza. La utilización del

analizador de tiempo de vuelo permitió una confirmación adicional de los positivos de PAHs

determinados mediante el triple cuadrupolo.

En la segunda parte de la Tesis se han explorado las posibilidades que ofrece el

analizador de tiempo de vuelo desde el punto de vista cualitativo y confirmativo de

contaminantes orgánicos mediante técnicas de screening. Concretamente se realizaron

estudios sobre muestras de sales marinas, salmueras y aguas medioambientales. La

utilización del analizador de tiempo de vuelo ha permitido la realización de diferentes

enfoques de análisis screening, bien con analitos preseleccionados (análisis target), analitos

no seleccionados previamente (análisis non-target) y analitos seleccionados post

adquisición con equipos de MS (análisis post-target).

Por último, las metodologías analíticas disponibles se aplican al estudio de la posible

bioacumulación de PAHs, OCs y PBDEs en los productos de la acuicultura marina, mediante

un estudio experimental que incluye un ciclo completo de engorde de doradas (Sparus

aurata L.) en el marco del Proyecto AQUAMAX (Sustainable Aquafeeds to Maximise the

Health Benefits of Farmed Fish for Consumers. www.aquamaxip.eu). Las muestras

consideradas en este estudio fueron materias primas de origen vegetal y animal, piensos

con diferente composición y filetes de dorada cultivada.

El trabajo desarrollado en esta Tesis Doctoral ha permitido concluir que la estrategia

a seguir depende principalmente del objetivo que se desea alcanzar, considerando el uso

de una instrumentación u otra en función de las necesidades requeridas.

En esta Tesis Doctoral se ha desarrollado y aplicado metodología analítica altamente

especializada con instrumentación avanzada basada en el acoplamiento GC-MS. La

metodología desarrollada presenta excelentes características analíticas y resulta de interés

tanto en análisis de rutina de estos contaminantes como para la investigación y búsqueda

de contaminantes de creciente interés, tanto en el mundo alimentario como en el campo

medioambiental.

ÍNDICE GENERAL

CAPÍTULO 1. INTRODUCCIÓN GENERAL 1

1.1 Introducción………………………………………………………………………………………………... 3

1.2 Determinación analítica de contaminantes orgánicos en muestras complejas.. 9

1.2.1 Determinación de POPs en muestras con elevado contenido graso……. 9

1.2.2 Determinación de contaminantes orgánicos en muestras acuosas y

muestras salinas………………………………………..………………………………………..

13

1.3 Cromatografía de gases……………………………………………………………………………… 14

1.4 Espectrometría de masas…………...…………………………………………………………………. 14

1.5 Acoplamiento de la cromatografía de gases a espectrometría de masas…….… 20

1.6 Aplicación de GC-MS para estudios screening de contaminantes orgánicos en

muestras ambientales y de salud pública ……………………………………………………….

22

1.7 Aplicación de GC-MS para estudios de bioacumulación…………………………...….... 23

1.8 Referencias…………………………………………………………………………………………………….. 24

CAPÍTULO 2. OBJETIVOS 31

CAPÍTULO 3. PLAN DE TRABAJO 35

CAPÍTULO 4. DETERMINACIÓN DE CONTAMINANTES ORGÁNICOS PERSISTENTES EN

MUESTRAS CON ALTO CONTENIDO GRASO 41

4.1 Introducción……………………………………….………………………………………………………… 43

4.2 Desarrollo de metodología analítica para la determinación de hidrocarburos

policíclicos aromáticos mediante GC-(QqQ)MS/MS y GC-TOF MS….………………..

47

4.2.1 Introducción.......................................................................................... 49

4.2.2 Artículo científico 1...............................................................................

A reliable analytical approach based on gas chromatography coupled

to triple quadrupole and time of flight analyzers for the determination

and confirmation of polycyclic aromatic hydrocarbons in complex

matrices from aquaculture activities

Rapid Commun. Mass Spectrom. 23 (2009) 2075-2086

53

4.2.3 Discusión de los resultados del artículo científico 1…………………………. 81

4.2.4 Referencias…………………………………………………………….…………………………. 85

4.3 Metodología analítica para la determinación de difenil éter polibromados en

muestras complejas………………………………………………………………..…….………………..

89

4.3.1 Introducción.......................................................................................... 91


Gas chromatography–mass spectrometric determination of

polybrominated diphenyl ethers in complex fatty matrices from

aquaculture activities

Anal. Chim. Acta 664 (2010) 190–198

93


4.3.4 Referencias…………………………………………………………….…………………………. 126

CAPÍTULO 5. DESARROLLO DE METODOLOGÍAS SCREENING BASADAS EN

CROMATOGRAFÍA DE GASES ACOPLADA A ESPECTROMETRÍA DE MASAS CON

ANALIZADOR DE TIEMPO DE VUELO PARA LA IDENTIFICACIÓN DE CONTAMINANTES

ORGÁNICOS 129

5.1 Introducción……………………………………….………………………………………………………… 131

5.2 Screening de contaminantes orgánicos en muestras de sal y agua de

mar…………………………………………..……………………………......………….…….………………..

135

5.2.1 Introducción.......................................................................................... 137


Non-target screening of organic contaminants in marine salts by gas

chromatography coupled to high-resolution time-of-flight mass

spectrometry

Submitted to Talanta

139


5.2.4 Referencias……………………………………………………..……….…………………………. 167

5.3 Desarrollo de metodología analítica para la identificación de OPEs en

muestras medioambientales…………………………………………………..…….………………..

171

5.3.1 Introducción............................................................................................ 173

5.3.2 Artículo científico 4.................................................................................

Investigation of OPEs in fresh and sea water, salt and brine samples by

GC-TOF MS.

Submitted to Analytical and Bioanalytical Chemistry

175

5.3.3 Discusión de los resultados del artículo científico 4……………………………. 195

5.3.4 Referencias…………………………………………………………….…………………………… 197

CAPÍTULO 6. ESTUDIO DE LA BIOACUMULACIÓN DE CONTAMINANTES ORGÁNICOS

PERSISTENTES EN PRODUCTOS DE LA ACUICULTURA MARINA. 199

6.1 Introducción……………………………………….………………………………………………………… 201

6.2 Estudio de la concentración y bioacumulación de hidrocarburos policíclicos

aromáticos en doradas procedentes de la acuicultura…………………………………….

205

6.2.1 Introducción.......................................................................................... 207


Bioaccumulation of Polycyclic Aromatic Hydrocarbons in Gilthead Sea

Bream (Sparus aurata L.) Exposed to Long Term Feeding Trials with

Different Experimental Diets.

Arch Environ Contam Toxicol 59 (2010) 137–146

211

6.2.3 Discusión de los resultados del artículo científico 5……………………………. 233

6.2.4 Referencias…………………………………………………………….…………………………... 235

6.3 Estudio de la bioacumulación de contaminantes organoclorados en doradas

procedentes de la acuicultura……………………………………………….…….………………….

237

6.3.1 Introducción.......................................................................................... 239

6.3.2 Artículo científico 6.................................................................................

Effects of fish oil replacement and re-feeding on the bioaccumulation

of organochlorine compounds in gilthead sea bream (Sparus aurata L.)

of market size.

Chemosphere 76 (2009) 811-817

243

6.3.3 Discusión de los resultados del artículo científico 6…………………….…….. 267

6.3.4 Referencias…………………………………………………………….…………………………… 270

CAPITULO 7. CONCLUSIONES…………….……...................................................................... 273

Artículos científicos presentados en la Tesis………………………………………………………………. 277

Artículos científicos relacionados con la Tesis……..……………………………………………..……… 279

Índice de acrónimos

ASE Extracción acelerada con disolventes

BaPE Equivalentes tóxicos del Benzo(a)Pireno

BMF Biomagnificación

CI Ionización química

DDD Diclorodifenildicloroetano

DDE Diclorodifenildicloroetileno

DDT Diclorodifeniltricloroetano

ELL Extracción líquido-líquido

EI Ionización electrónica

EtOH Etanol

EU Unión europea

GC Cromatografía de gases

GC-MS Cromatografía de gases acoplada a espectrometría de masas

GPC Cromatografía por permeación en gel

H2SO4 Ácido sulfúrico

HPLC Cromatografía líquida de alta resolución

ICP Fuente de plasma de acoplamiento inductivo

IT Analizador de trampa de iones

Kow Constante de reparto octanol/agua

KOH Hidróxido potásico

LC Cromatografía líquida

LC-MS Cromatografía líquida acoplada a espectrometría de masas

LOD Límite de detección

LOI Límite de identificación

LOQ Límite de cuantificación

m/z Relación masa/carga

MAE Extracción asistida por microondas

MeOH Metanol

MRM Monitorización de transiciones seleccionadas (multiple reaction monitoring)

MS Espectrometría de masas

MS/MS Espectrometría de masas en tándem

NaCl Cloruro sódico

NaOH Hidróxido sódico

NCI Ionización química negativa

NL Pérdidas neutras

NPLC Cromatografía líquida en fase normal

OCs Organoclorados

OCPs Pesticidas organoclorados

OPs Organofosforados

OPE Éster organofosforado

PAHs Hidrocarburos policíclicos aromáticos

PBDEs Difenil éter polibromados

PCBs Bifenilos policlorados

PCDDs Dibenzodioxinas

PCDFs Dibenzofuranos

PCI Ionización química positiva

PLE Extracción líquida presurizada

POPs Contaminantes orgánicos persistentes

Q Analizador de masas cuadrupolar

QqQ Analizador de masas triple cuadrupolo

RSD Desviación estándar relativa

SIM Monitorización de iones seleccionados (selected ion monitoring)

SIR Monitorización de iones seleccionados (selected ion recording)

SPE Extracción en fase sólida

SPME Microextracción en fase sólida

SRM Monitorización de transiciones seleccionadas (selected reaction

monitoring)

TEF Factor de equivalencia tóxica

TIC Cromatograma de iones totales (total ion chromatogram)

TOF Analizador de masas de tiempo de vuelo

USEPA Agencia americana de protección del medioambiente

CAPÍTULO 1

INTRODUCCIÓN GENERAL

Capítulo 1. Introducción general

3

1.1. Introducción

En las últimas décadas, el aumento progresivo de la presencia de algunos

contaminantes orgánicos persistentes (POPs) y la aparición de nuevos contaminantes,

producidos por la actividad humana en el medio ambiente, ha supuesto una creciente

preocupación acerca del efecto que estos pueden tener sobre los ecosistemas y los

organismos vivos.

La presencia de estos contaminantes en organismos vivos hace que sea viable su

ingesta y posterior asimilación por organismos de niveles tróficos superiores. Estos

contaminantes suponen una amenaza importante para la industria alimentaria. Los

organismos de control oficial deben asegurar la inocuidad de los mismos, proporcionando

las máximas garantías de calidad y seguridad alimentaria.

Se conocen como contaminantes orgánicos aquellas sustancias orgánicas con

características definidas de toxicidad, carcinogeneidad y mutagenicidad presentes en

cualquiera de los compartimentos medioambientales (Clement et al., 1999). Entre los

diferentes tipos de contaminantes orgánicos existentes en la actualidad, posiblemente son

los plaguicidas y los POPs los que han recibido mayor atención (Eljarrat et al., 2003). Su

demostrada toxicidad y elevado uso en todo el planeta, así como las posibilidades de

contaminación de distintos tipos de muestras (suelos, aguas, aire, productos alimenticios,

etc.) son, sin duda, las causas de su carácter prioritario como contaminantes a controlar.

Además, presentan un importante riesgo de bioacumulación a lo largo de la cadena trófica,

llegando a alcanzar, en ocasiones, concentraciones alarmantes en los últimos eslabones de

la misma (Hites et al., 2004). Por otro lado, algunas de estas sustancias muestran una


4

elevada persistencia en el medio ambiente. Esto se debe a su lenta degradabilidad, por lo

que pueden llegar a detectarse residuos hasta varios años después de haber aplicado el

producto.

La preocupación por los efectos nocivos derivados de estos contaminantes es una

constante de nuestra sociedad actual. Asimismo, el número de compuestos orgánicos de

toxicidad reconocida usados en la actualidad es muy elevado y su composición química muy

variada. Por ello, los efectos tóxicos y los problemas ambientales derivados de su utilización

son complejos y muy variados. Así, resulta necesaria una prevención adecuada, con el

objeto de proteger la salud de toda la población en general.

La problemática de estos contaminantes en el mundo sigue siendo en la actualidad

uno de los temas más desconocidos por diversos motivos. La dificultad de analizar estos

compuestos a los bajos niveles de concentración exigidos por la legislación y la carencia de

listas prioritarias a controlar en distintas áreas de interés, son quizás las razones que, en

principio, pueden indicarse como más importantes. La contaminación orgánica conlleva una

problemática que, a menudo, sobrepasa fronteras, lo que implica necesariamente la

adopción de medidas legales a nivel internacional como las que se citan a continuación:

• CONVENIO DE BASILEA (22 de Marzo de 1989, entró en vigor en 1992).

Sobre el control de los movimientos transfronterizos de los desechos peligrosos y

su eliminación.

• CONVENIO DE ROTTERDAM (11 de Septiembre de 1998).

Para la aplicación del procedimiento relativo a ciertos plaguicidas y químicos

peligrosos objeto de comercio internacional.

• CONVENIO DE ESTOCOLMO (23 de Mayo de 2001, ratificado 17 de mayo de 2004).

El objetivo del Convenio es proteger la salud humana y el medio ambiente frente

a los POPs.


5

• REACH – ‘The New EU Chemicals Legislation’ (29 de Octubre de 2003).

Los objetivos son mejorar la protección de la salud humana y del medio ambiente,

manteniendo la competitividad y promoviendo la innovación de la industria

química en la Unión Europea (EU).

Un ejemplo claro de la problemática de la contaminación es el caso de la acuicultura.

El estudio de contaminantes orgánicos tales como los OCs, pesticidas organoclorados

(OCPs), los bifenilos policlorados (PCBs), hidrocarburos policíclicos aromáticos (PAHs),

retardantes de llama como los difenil éter polibromados (PBDEs), dioxinas y compuestos

similares, muestran valores divergentes entre animales salvajes y los que se crían en

granjas, y a su vez, entre los que se crían en América y Europa (Hites et al., 2004; Serrano et

al., 2008). Atribuir este fenómeno a problemas de tipo medioambiental sería difícil de

asumir. Entre otras razones, porque significaría que las condiciones ambientales de los

animales criados en cautividad son mucho peores que las de las aguas donde crecen en

libertad. Aun admitiendo que hubiera contaminación ambiental, que ciertamente existe en

las piscifactorías de dorada, lubina o salmón por ejemplo, y por supuesto en los ambientes

marinos libres, hay que considerar en paralelo el fenómeno de la bioacumulación. En los

animales salvajes, probablemente debido a que su dieta en los primeros estadios se basa en

insectos y otros animales que no han llegado a acumular niveles importantes de estos

contaminantes, la presencia de compuestos OCs en sus tejidos grasos no es alta a no ser

que se dé una exposición aguda. Los mayores niveles sólo suelen presentarse en los

animales de mayor edad.

Consecuentemente, si en las piscifactorías se detectan contaminantes OCs, que

típicamente se acumulan en hígado y partes grasas del animal, y no hay exposición aguda o

mecanismos que propicien la bioacumulación, la única vía que explicaría su presencia son

los piensos. La vía de contaminación mediante piensos en producción animal es bien

conocida. Piensos en mal estado, tratados inadecuadamente o enriquecidos con materiales

que se han demostrado poco apropiados, están en el origen de más de una crisis

alimentaria. Ejemplos recurrentes de ello es el uso irregular e indiscriminado de antibióticos


6

u otros medicamentos a través de los piensos, la todavía candente crisis de las vacas locas y

la pérdida de confianza en los productos cárnicos alemanes de principios de año como

consecuencia del escándalo generado por la presencia de contaminantes en piensos

animales.

A diferencia de la alimentación natural, los animales de piscifactoría se alimentan con

harinas que se obtienen a partir de restos de pescado, entre otros. Entre los componentes

importantes se encuentra el aceite de pescado, ya que de otra manera, difícilmente se

podría conseguir una composición de calidad de la grasa del pescado cultivado, y no

encontraríamos los efectos beneficiosos del consumo reiterado de su grasa (aceites

esenciales omega-3 y omega-6) sobre las enfermedades cardiovasculares. Es en esta

fracción grasa donde nos encontraríamos el principal problema. La mayor parte de los POPs

analizados, tales como los PAHs, PBDEs, PCBs y algunos OCs, son acumulables en la fracción

grasa de las muestras analizadas; harinas y aceites de pescado, piensos y filetes de pescado

entre otros. Si se utilizan especies menores para la elaboración de harinas de pescado, o

incluso restos de otras especies grasas de mayor tamaño, se aprecia, como han puesto de

manifiesto numerosos estudios, un incremento de la concentración de los niveles de la

contaminación química (Dórea et al., 2006).

Estos hechos ponen de manifiesto una verificación de este principio ecológico. Esta

situación, que si bien no parece de un gran peligro, si que presenta un problema que

requiere solución, siendo ésta un adecuado y riguroso control de las materias primas

empleadas en la elaboración de los piensos para la alimentación de los peces (Newsletters

from Aquamax Project, http://www.aquamaxip.eu).

Como ejemplo, la Organización Mundial de la Salud considera como máximos

aceptables concentraciones de dioxinas comprendidas entre 1-4 pg/Kg peso vivo del

consumidor al día. Por tanto, un individuo adulto de 70 Kg de peso podría ingerir 280

pg/día. Si vemos que los máximos detectados son de 3 pg/g de pescado, se podrían ingerir

hasta 90 g de salmón al día (Organización mundial de la salud, 2001,

http://www.who.int/inf-pr-2001/en/state2001-01.html).


7

Efectivamente, las cifras no pueden ser consideradas tan alarmantes desde este

punto de vista. Sin embargo, si en el futuro no se consideran medidas de control que

consigan reducir las concentraciones actuales, sí que podríamos vernos abocados a una

situación de peligro difícilmente controlable.

Con el fin de evitar la contaminación en alimentos derivados de origen animal y

vegetal se han de vigilar y controlar rigurosamente las etapas de producción,

almacenamiento y distribución de los piensos hasta el momento de su consumo. Para ello

se requieren técnicas de análisis avanzadas para la identificación de estos agentes nocivos.

Estas técnicas han de ser capaces de proporcionar resultados fiables para la determinación

de estos contaminantes en cortos periodos de tiempo y de manera fiable. La elevada

toxicidad de los contaminantes presentes en el medio ambiente obliga además a disponer

de metodología analítica fiable, rápida y suficientemente sensible que permita el control de

todo tipo de contaminantes en el medio ambiente.

Otra problemática sobre contaminación medioambiental lo constituye la

contaminación de los recursos hídricos. Mares, ríos y aguas superficiales han sido y siguen

siendo en la actualidad el destino de gran cantidad de residuos, deshechos y vertido de

compuestos químicos que en muchas ocasiones presentan toxicidad. El origen de esta

posible contaminación procede de la industria, de las zonas urbanizadas, de actividades de

la agricultura y ganadería, entre otros. Las aguas superficiales, subterráneas e incluso los

mares contienen una elevada carga de contaminantes de origen antropogénico, muchos de

los cuales tienen una alta persistencia en el medio ambiente y la capacidad de

bioacumularse en las redes tróficas acuáticas. Como ejemplo de compuestos

antropogénicos podemos citar a los OCs, PCBs, PBDEs, PAHs, entre otros. Estos

contaminantes han sido ampliamente estudiados debido a la gran preocupación que han

provocado por su peligrosidad para el medio ambiente y la salud pública. En este contexto,

la aparición de nuevos contaminantes en muestras ambientales, como es el caso de los

compuestos farmacéuticos y de cuidado personal, ha despertado el interés hasta el punto

de ser considerados como un problema medioambiental aún por determinar (Daughton et

al 1999; Ternes et al., 2004; Muñoz et al., 2008).


8

La contaminación de las aguas medioambientales, y en concreto de las aguas

marinas, tiene como consecuencia la aparición de contaminantes en productos de consumo

humano. En este sentido las salmueras y sales marinas de uso alimentario, proceden de la

evaporación del agua de agua de mar y la cristalización de la sal presente en ella. Muchos

de los contaminantes presentes en el agua de mar que llegan a las salinas, enclaves muy

expuestos y sensibles a la contaminación, pueden permanecer tanto en la sal como en las

salmueras después del proceso de evaporación. Este hecho se convierte en una amenaza

para la seguridad alimentaria, ya que gran cantidad de alimentos de consumo humano

contienen sal. También las salmueras constituyen cierto riesgo por ser ampliamente

utilizadas en alimentación, pesquerías y productos de la acuicultura.

Resulta de gran interés el análisis y control de sales y salmueras ya que, hasta hoy, y

bajo nuestro conocimiento, existen pocos trabajos o estudios publicados acerca de la

presencia de contaminantes en este tipo de muestras. Su estudio contribuye a un mayor

conocimiento y percepción en materia de salud pública y seguridad alimentaria.

El análisis de contaminantes orgánicos en muestras medioambientales sufrió un

excelente impulso con la aparición de la cromatografía de gases (GC). El uso correcto de

esta técnica permitió la separación de un gran número de compuestos. Posteriormente, la

introducción de la GC capilar a finales de los años 70 y la posibilidad de disponer de

sistemas de detección de espectrometría de masas (MS) selectivos y fiables, contribuyó a la

emergente popularidad obtenida por los procedimientos basados en GC. A principios de los

años 80, los primeros sistemas de cromatografía líquida acoplada a espectrometría de

masas (LC-MS) empezaron a comercializarse. Gracias a la técnica LC-MS se pudo ampliar el

análisis de contaminantes orgánicos ya que además de ofrecer una excelente separación,

fue posible llevar a cabo el análisis de analitos que por características físico-químicas

(termolabilidad, volatilidad, polaridad) eran más problemáticos mediante GC-MS.

En la actualidad, la determinación de contaminantes orgánicos se lleva a cabo

mayoritariamente mediante técnicas cromatográficas. El análisis de compuestos orgánicos

a niveles de traza, no se puede llevar a cabo con un análisis directo de la muestra, sino que


9

es necesario realizar un pretratamiento de esta que permita su introducción en el equipo

de medida.

1.2. Determinación analítica de contaminantes orgánicos en muestras complejas.

La determinación de contaminantes orgánicos en muestras complejas con alto

contenido graso, en aguas medioambientales, muestras de sales o salmueras, es una tarea

complicada. Resulta ser un procedimiento especialmente delicado por tratarse de muestras

con gran cantidad de interferentes que complican la determinación instrumental, tanto

cualitativa como cuantitativa. Por ello, es imprescindible la aplicación de técnicas eficaces

que permitan la obtención de extractos limpios compatibles con los sistemas de medida,

para proporcionar resultados fiables y representativos. Otro factor que complica la

determinación es la baja concentración de estos contaminantes en las muestras de análisis,

lo que hace necesario la aplicación de técnicas sensibles y selectivas.

Atendiendo a los compuestos a determinar y la tipología de las muestras, se deberá

seguir un procedimiento específico para cada caso con el fin de obtener resultados

satisfactorios. Cabe destacar, que en todas las experiencias de esta Tesis se prestó especial

atención a las guías SANCO (SANCO/10684/2009; SANCO/3131/2007) para la validación y

control de calidad de la metodología analítica desarrollada.

1.2.1. Determinación de POPs en muestras con alto contenido graso

El primer paso en la determinación de POPs es la extracción de los compuestos de

interés de la matriz. Posteriormente, dependiendo del origen de la muestra y de su

complejidad, serán necesarias una o varias etapas de purificación y eliminación de

interferentes con el fin de obtener resultados satisfactorios y fiables.

En nuestro caso, en muestras de matriz compleja como los filetes de pescado,

piensos o materias primas animales o vegetales, la extracción sólido-líquido mediante un


10

sistema de reflujo resulta ser una excelente técnica para extraer adecuadamente los

compuestos orgánicos de la matriz sólida y transferirlos a la disolución orgánica (Serrano et

al., 2003). El contacto directo del disolvente orgánico extractante con la matriz,

generalmente n-hexano o diclorometano, hace que los analitos afines al disolvente sean

atraídos por éste consiguiéndose una extracción más efectiva. El efecto de la temperatura

en la etapa de reflujo facilita la disgregación de la muestra mejorando el contacto del

disolvente con la matriz. De esta manera también se obtiene una extracción más óptima. Se

debe tomar especial precaución con aquellos compuestos de elevada volatilidad, teniendo

en cuenta la temperatura en la que se realiza el reflujo para evitar pérdidas. Los principales

inconvenientes a destacar en esta técnica de reflujo son, un elevado consumo de

disolventes orgánicos y tiempo de extracción, que alcanza en muchos casos las cuatro

horas. Asimismo, presenta dificultad de automatización.

Otro sistema de extracción ampliamente conocido es el Soxhlet. Esta técnica tiene un

fundamento similar al reflujo, siendo muy efectiva y ampliamente utilizada para la

extracción de compuestos orgánicos en matrices complejas. El uso de disolventes con

elevada afinidad por los analitos permite una óptima extracción de estos de matrices

complejas, lo que hace que estas técnicas sean frecuentemente utilizadas en numerosas

aplicaciones (De Koning et al., 2009).

Metodologías de extracción como la extracción líquida presurizada (PLE), extracción

acelerada con disolventes (ASE), extracción asistida por microondas (MAE), microextracción

en fase sólida (SPME) y extracción con fluidos supercríticos (Schantz et al., 2006; Fidalgo-

Used et al., 2007), aceleran el tiempo de extracción y facilitan la manipulación de la

muestra después de la extracción (van Leeuwen et al., 2008; Björklund et al., 2006).

Después de la etapa de extracción, los analitos objeto de estudio se encuentran

asociados a la matriz y conviene realizar alguna etapa de purificación lo más específica

posible que garantice la eliminación de los interferentes. De esta forma, será viable la

concentración de los analitos en el extracto y así poder conseguir una metodología más

sensible y selectiva. Todos aquellos compuestos, contenidos en la muestra, que no son


11

objetivo de análisis, pueden llegar a enmascarar nuestros analitos y disminuir la

sensibilidad, selectividad y especificidad de nuestra metodología. Los interferentes

presentes en este tipo de muestras son generalmente compuestos de elevado peso

molecular, principalmente lípidos, que dificultan el análisis.

La purificación o “clean-up” de una muestra es una de las etapas más importantes en

el análisis de muestras complejas. Para la determinación de compuestos ácido-resistentes

como los PBDEs, los PCBs y algunos OCs, el tratamiento de la muestra con ácido sulfúrico

garantiza la eliminación de más del 90% del contenido lipídico de la muestra (Serrano et al.,

2003). En el caso de los compuestos ácido-lábiles como los PAHs, es necesario proponer

rutas de purificación alternativas (Fernández-González et al., 2008 a, b). Una buena solución

para la eliminación de interferentes sería la realización de una purificación mediante una

etapa de saponificación. En esta etapa, la fase lipídica de las muestras se transforma en

jabón y los analitos orgánicos, PAHs en este caso, pasan a la fase apolar (Perugini et al.,

2007).

Otra técnica de purificación ampliamente utilizada en los laboratorios de rutina para

la determinación de contaminantes orgánicos ha sido la extracción en fase sólida (SPE) (Picó

et al., 2007). Esta técnica supone una alternativa rápida y eficaz comparada con la etapa

tradicional de extracción líquido-líquido (ELL). Comparando ambas técnicas, la SPE es una

técnica que precisa una menor utilización de disolventes orgánicos, un volumen menor de

muestra, evita además la formación de emulsiones y, lo más importante, permite

automatizar el proceso y así realizar la purificación de más de una muestra al mismo

tiempo.

El procedimiento de SPE se debe optimizar con el fin de obtener resultados

satisfactorios para la eliminación de interferentes y concentración de nuestros analitos. La

selección de la fase estacionaria, el disolvente de elución, el estudio del patrón de elución y

volumen de ruptura (breakthrough) son variables críticas a optimizar.

Para purificar muestras con contenido graso mediante SPE se suele trabajar en fase

normal. En esta modalidad, se pretende eluir los analitos mientras que los interferentes


12

(generalmente lípidos) son retenidos en la fase estacionaria y eluidos posteriormente en

otra fracción diferente a la de los analitos de interés (Meloan et al., 1999). Una técnica de

purificación muy similar a la anterior es la cromatografía líquida en fase normal (NPLC),

también empleada para la purificación de muestras con contenido lipídico (Serrano et al.,

2003). Las fases estacionarias comúnmente empleadas son de silica y Florisil.

Por último, cabe destacar también la cromatografía por permeación en gel (GPC). La

GPC es una técnica ampliamente utilizada en laboratorios de análisis de pesticidas en

muestras con contenido graso. Esta técnica permite la separación casi completa entre los

lípidos y los pesticidas (Tindle et al., 1972; Griffitt et al., 1974). La GPC permite la

purificación de extractos grasos o la separación de fracciones lipídicas. La separación se

basa en el tamaño molecular de los compuestos. El sistema GPC consiste básicamente en

una columna rellena de un polímero poroso en forma de bolas. Al introducir el extracto

lipídico, las moléculas más pequeñas se retienen más en la columna puesto que deben

recorrer mayor camino a través de los poros del relleno polimérico mientras que las

moléculas grandes eluirán primero. Una buena elección del relleno de la columna de GPC y

del disolvente empleado para la elución permite generalmente una separación de los

pesticidas de la fracción lipídica en la mayoría de los casos.

A pesar de disponer de técnicas excelentes para la purificación de muestras grasas, la

correcta determinación de POPs a niveles de traza resulta en muchas ocasiones insuficiente

mediante una única etapa de purificación. Generalmente, después del primer clean-up

todavía quedan interferentes por lo que es necesario realizar un segundo tratamiento de la

muestra que nos permita cuantificar los analitos a niveles bajos de concentración con una

buena sensibilidad (Serrano et al., 2003).

En nuestro caso, para la determinación de PAHs, el tratamiento de las muestras

consistió en una saponificación y posterior SPE con Florisil mientras que para la

determinación de PBDEs, se realizó una extracción con reflujo seguida de tratamiento con

ácido y posterior SPE con Florisil obteniendo resultados satisfactorios en ambos casos. Los

procedimientos desarrollados permiten llevar a cabo la purificación de muestras de la


13

acuicultura con elevado contenido graso, pudiendo inyectar los extractos resultantes

directamente en el sistema GC-MS.

1.2.2. Determinación de contaminantes orgánicos en muestras acuosas y muestras

salinas.

Como se ha comentado anteriormente, la SPE es una de las técnicas más populares y

ampliamente aceptadas para el análisis de contaminantes orgánicos. Hoy en día, el

tratamiento de muestra mediante SPE junto al potencial del acoplamiento GC-MS ha sido

ampliamente utilizado para la determinación de contaminantes orgánicos en muestras

ambientales (Rubio y Pérez-Bendito, 2009; Pitarch et al., 2010).

Para el análisis de contaminantes orgánicos en muestras de disoluciones acuosas y

salinas, se emplean preferentemente cartuchos de SPE en los que se trabaja en fase

inversa. La purificación de las muestras se realizará con cartuchos de SPE de fase

estacionaria apolar, entre las que destacan las fases C18, C8, fenilo, entre otras. Los analitos

de interés son retenidos al pasar la muestra por la fase estacionaria y posteriormente estos

analitos se eluyen mediante un disolvente apolar.

En estos casos, la SPE es una técnica económica, rápida, de fácil automatización y que

no emplea gran cantidad de disolventes, además de permitir una buena concentración del

extracto de la muestra. Únicamente pasarán al extracto concentrado final aquellas

sustancias apolares afines a la fase estacionaria y al disolvente orgánico escogido para la

elución. De esta manera se obtiene un tratamiento suficientemente selectivo de los

analitos objeto de estudio permitiendo alcanzar excelente sensibilidad y niveles de

cuantificación bajos.

En la presente investigación se realizó la técnica de SPE mediante cartuchos C18

consiguiendo una eficiente purificación y concentración de las muestras acuosas y salinas.


14

1.3. Cromatografía de gases

La GC es una técnica de separación ampliamente utilizada en análisis de

contaminantes orgánicos en muestras ambientales (Poster et al., 2006; Van Leeuwen et al.,

2008). Los objetivos de una técnica cromatográfica en columna son la separación de las

diferentes especies químicas en zonas o bandas que se mueven a diferente velocidad, las

cuales se visualizan en un cromatograma mediante un sistema de detección a la salida de la

columna. La diferente distribución de los componentes de la muestra entre una fase

estacionaria y una fase móvil junto a las diferentes velocidades de desplazamiento de los

componentes al ser arrastrados por la fase móvil a través de la fase estacionaria serán

factores que determinaran la cromatografía.

Pese a la importancia de los factores que intervienen en una buena separación en

GC, su discusión teórica escapa de los objetivos de la presente Tesis, pudiéndose consultar

bibliografía sobre los aspectos teóricos de la GC (Meloan et al., 1999; Cela et al., 2002; Van

Leeuwen et al., 2008) así como parámetros básicos indispensables a optimizar a la hora de

obtener unos resultados analíticos satisfactorios (Kitson et al., 1996; McMaster et al.,

1998).

Cabe destacar que todos los trabajos que se presentan en los siguientes capítulos

están basados en GC debido a la volatilidad y carácter hidrófobo que presentan los analitos

de estudio.

1.4. Espectrometría de masas

La técnica de MS ha supuesto una mejora en la medida de la masa molecular de

compuestos y átomos mediante la conversión de estos en iones cargados. La popularidad

de esta técnica radica fundamentalmente en su especificidad, sensibilidad, información

estructural, entre otras (Barker et al., 1998). Esta técnica se basa principalmente en

diferentes procesos de ionización, separación y detección (Chhabil et al., 2007). Estos

procesos transcurren en condiciones de alto vacío lo que permite que los iones se muevan


15

libremente sobre un espacio sin interacciones con otras especies. Las posibles colisiones

producidas en un analizador de masas, podrán dar lugar a fragmentaciones de los iones

moleculares pudiendo producir, mediante reacciones moleculares, diferentes especies

iónicas.

Los componentes de un MS se muestran en la Figura 1.1 y estos son los siguientes:

Figura 1.1. Componentes de un espectrómetro de masas.

1. Sistema de Inyección. Espacio reservado para la entrada de la muestra y su

transformación previa entrada a la fuente de ionización. En este proceso la

muestra entra al sistema a presión atmosférica y entra en el vacío generado por el

sistema.

2. Fuente de ionización: Espacio en el cual las moléculas neutras procedentes de la

muestra se ionizan y se mezclan con el gas de iones. La fuente de ionización

empleada cuando se realiza un acoplamiento GC-MS puede ser de dos tipos:

2.1. Ionización electrónica (EI): ionización mediante la cual un electrón

interacciona con la molécula arrancando un electrón de la misma. En el

proceso, la molécula queda con un exceso de energía que se disipa por

vibración, rotación, reordenación molecular o mediante fragmentación.

2.2. Ionización química (CI). Ionización más débil que la anterior. El fundamento

consiste en transferir la carga desde un gas reactivo (ej. Metano) a las

Corrientede vacio

Fuente de Ionización

Analizador Detector

Sistema de Inyección (ej. GC)

Procesamientodatos

Iones en fase gas Selección de iones Detección de iones

Introducción de la muestra Resultados, cromatograma, Espectro de masas.

Corrientede vacio

Fuente de Ionización

Analizador Detector

Sistema de Inyección (ej. GC)

Procesamientodatos

Iones en fase gas Selección de iones Detección de iones

Introducción de la muestra Resultados, cromatograma, Espectro de masas.


16

moléculas de analito. El exceso de energía sería mínimo y se produce una

escasa fragmentación a diferencia de la fuente EI. Su principal uso está

dirigido a potenciar la abundancia del ión molecular en aquellas sustancias en

las que su presencia en el espectro de EI es escasa o no aparece. Esta

ionización puede dar lugar a especies positivas (ionización química positiva,

PCI) o especies negativas (ionización química negativa, NCI).

3. Analizador de masas: El analizador es el responsable de separar los distintos

fragmentos en función de su relación m/z. Las características principales de un

analizador de masas son la resolución, sensibilidad, rango de masas y capacidad

de medida exacta de masas. Existen diferentes tipos de analizadores entre los que

podemos destacar el sector magnético, cuadrupolo (Q), triple cuadrupolo (QqQ),

trampa de iones (IT) y tiempo de vuelo (TOF). Como los trabajos realizados en esta

Tesis se basan en el desarrollo de metodologías analíticas mediante analizadores

QqQ y TOF, y en el último capítulo se presenta una aplicación con analizador IT,

este apartado se centra exclusivamente en estos analizadores (funcionamiento,

ventajas, inconvenientes…) ya que el resto se encuentra ampliamente descrito en

bibliografía (Barker et al., 1998; Chhabil et al., 2007).

3.1. Analizadores QqQ e IT: En los últimos años, la espectrometría de masas en

tándem (MS/MS) ha sido cada vez más utilizada para el análisis de

contaminantes orgánicos en matrices complejas. En esta Tesis se han

utilizado analizadores QqQ e IT para aplicaciones MS/MS. Gracias al

acoplamiento MS/MS se obtiene gran sensibilidad y selectividad en la

determinación de contaminantes orgánicos, dado que se minimizan en gran

medida las interferencias debidas a la matriz. MS/MS se refiere a la

combinación de dos puntos de análisis de masas que pueden darse tanto en

el tiempo como en el espacio. Hasta ahora, las aplicaciones y la popularidad

de esta técnica han ido creciendo paulatinamente (Santos et al., 2003;

Chhabil et al., 2007; Fernández-González et al., 2008b). Se le pueden atribuir

contribuciones en el campo de la elucidación de estructuras moleculares de


17

compuestos desconocidos, identificación de compuestos en matrices

complejas, elucidación de rutas de fragmentación y cuantificación de

compuestos en muestras reales. Las características de un MS/MS podrían

resumirse en los siguientes puntos:

− Proporciona información estructural.

− Eliminación de picos interferentes. Excelente selectividad.

− Mejora de la señal/ruido.

− Mejora de la sensibilidad.

− Aplicación a muestras de matriz compleja.

Por otra parte, los inconvenientes de esta técnica serían:

− Instrumentación más cara y compleja.

− Búsqueda en librerías no disponible.

En un analizador QqQ existen diferentes modalidades de trabajo puesto que

disponemos de dos analizadores (Figura 1.2). El primer analizador permite

Figura 1.2. Modos de trabajo en MS/MS.

Analizador 1(Q)SIR

Celda de Colisión(H)

Analizador 2(q)

Scan

Barrido de iones producto

Analizador 1(Q)

Scan


Analizador 2(q)SIR

Búsqueda de iones precursores

Analizador 1(Q)

Scan


Analizador 2(q)

Scan

Búsqueda de pérdidas neutras

Analizador 1(Q)SIR


Analizador 2(q)SIR

Selected Reaction monitoring (SRM)


18

trabajar en modo SIR (Selected ion Recording) o en modo “full scan”. En

modo SIR, se selecciona un ión producto específico, y sólo se mide éste. En

“full scan”, se adquiere un espectro de masas continuo en un rango definido,

a una velocidad de barrido adecuada. El segundo analizador permite

múltiples posibilidades: barrido de iones producto, búsqueda de iones

precursores, búsqueda de pérdidas neutras (NL), selected reaction

monitoring (SRM). En la modalidad SRM, se disminuye en gran medida el

ruido químico aumentando significativamente la relación S/N y observando

un notable aumento de la sensibilidad. Trabajando con transiciones

específicas en modo SRM se obtienen metodologías altamente selectivas

incrementando la confirmación de la identidad de los analitos.

En el analizador IT, la ruptura y aislamiento de los iones se produce en el

tiempo, mientras que para el analizador QqQ, este proceso se realiza en el

espacio, presentando ambos analizadores ventajas y desventajas. En

concreto, el QqQ está compuesto de tres cuadrupolos colocados

consecutivamente, de los cuales el primero y el tercero seleccionan los iones

de interés, y el segundo actúa como celda de colisión. Este segundo

cuadrupolo es el que permite la realización de MS/MS, puesto que el ciclo

ruptura-detección se realiza dos veces. Físicamente se aceleran los iones

precursores desde el primer cuadrupolo hacia el segundo en presencia de un

gas de colisión inerte (generalmente argón o helio), pudiendo obtenerse

distintas fragmentaciones.

Garrido Frenich et al., 2008 realizaron una comparación entre los

analizadores QqQ e IT acoplados a GC para la determinación de pesticidas en

alimentos de consumo. En general, los analizadores IT ofrecen mayor

información estructural que los QqQ además de una mayor robustez en las

medidas. Mejor sensibilidad y selectividad presenta el QqQ respecto al IT

para muestras más complejas, además de una mayor velocidad de scan,


19

permitiendo la inclusión de un mayor número de compuestos por método

analítico.

Para llevar a cabo una correcta confirmación de la identidad de los

compuestos detectados, en esta Tesis se seguirá el criterio de identificación

por puntos propuesto por la Unión Europea (Commission Decision

2002/657/EC). En nuestro caso, trabajando en MS/MS por cada ión precursor

se obtiene un punto de identificación (IP) y por cada ión producto 1,5 puntos.

Trabajando en modo SIR, por cada ión adquirido obtendremos 1 punto de

identificación. Para la confirmación de compuestos prohibidos es necesario

conseguir 4 IPs mientras que para compuestos legales 3 IPs (Hernández et al.,

2005).

3.2. Analizadores TOF. El analizador TOF es, en cuanto a fundamento, uno de los

analizadores más simples de los que se utilizan hoy en día. Se basa en la

medida del tiempo que tardan los iones generados y acelerados con igual

energía en la fuente de iones, en alcanzar un electrodo colector situado a una

distancia prefijada. Como los iones poseen la misma energía pero diferentes

masas, éstos alcanzarán el colector a diferentes tiempos, dependiendo de su

masa, carga y energía cinética. Los analizadores TOF destacan por su elevada

resolución (~7000 FWHM) y exactitud de masa.

Estos analizadores son capaces de proporcionar información espectral

completa de alta resolución permitiendo adquirir gran cantidad de

información química en un único análisis. También permite aplicaciones de

búsqueda de compuestos desconocidos mediante técnicas de screening.

Además, el analizador TOF es capaz de analizar simultáneamente todos los

analitos dentro de un rango de masas predeterminado, no así los

analizadores de masas como el Q o IT, donde las masas son preseleccionadas,

aisladas y analizadas secuencialmente. Gracias a estas posibilidades, la


20

combinación de GC junto a un analizador TOF ofrece una excelente

sensibilidad en la adquisición de un espectro completo.

Como se ha comentado anteriormente, para la confirmación de la

identidad de los compuestos detectados, se seguirá el criterio de

identificación por puntos propuesto por la Unión Europea (Commission

Decision 2002/657/EC).

4. Detector: Los iones generados llegan al detector que al recibir el impacto

producido por las partículas cargadas emite electrones. Estos electrones son

acelerados hacia un dínodo el cual emite varios electrones al recibir el impacto de

cada electrón. Este proceso se repite varias veces hasta obtenerse una cascada de

electrones que llega al colector lográndose una corriente fuertemente

amplificada. La corriente obtenida puede amplificarse de nuevo por

procedimientos electrónicos y se lleva a un sistema de procesamiento.

5. Procesamiento de datos: Los espectros de masas se almacenan y procesan en un

ordenador. Es muy frecuente la construcción de bibliotecas de espectros de

masas que permiten la identificación de compuestos químicos por comparación.

1.5. Acoplamiento cromatografía de gases a espectrometría de masas.

El acoplamiento GC-MS combinando la técnica de separación y la detección e

identificación de los componentes de una mezcla o matriz compleja, es una técnica

ampliamente utilizada en laboratorios analíticos de investigación. GC-MS es una de las

técnicas más atractivas y potentes para el análisis rutinario de contaminantes orgánicos

debido a que ofrece gran sensibilidad y selectividad. En un sistema GC-MS, las muestras son

introducidas al GC pasando éstas a estado gas y mezclándose con un gas inerte que es

empujado por una alta corriente de vacío hacia el analizador de MS (Figura 1.3).


21

Horno de temperaturaprogramable

Inyector

Gas: He, N2, H2

Columna capilar

Espectrómetro de masas

Figura 1.3. Sistema GC-MS.

Una vez producida la separación, los compuestos separados van llegando al MS

gracias a la corriente de vacío. Los compuestos se ionizan y los iones son seleccionados en

relación a su relación m/z (McMaster et al., 1998).

GC-MS es una técnica popular, potente, relativamente económica y frecuentemente

utilizada para la realización de estudios analíticos ambientales. La GC es una técnica

separativa que tiene la cualidad de conseguir la separación de mezclas muy complejas. Una

vez separados los componentes individuales de una muestra problema por GC, el único

dato que disponemos para la identificación de cada uno de ellos es el tiempo de retención

de los correspondientes picos cromatográficos. Este dato no es suficiente para una

identificación, sobre todo cuando analizamos muestras con un número elevado de

componentes, que es el caso frecuente. De ahí la importancia del acoplamiento GC-MS.

Como se ha comentado anteriormente, los analizadores MS pueden identificar de

manera casi inequívoca cualquier sustancia pura, pero normalmente no es capaz de

identificar los componentes individuales de una mezcla sin haberlos separado previamente.

Esto es debido a la extrema complejidad del espectro obtenido por superposición de los

espectros particulares de cada componente. Por tanto, la asociación de las dos técnicas, da

lugar a una técnica combinada GC-MS que permite la separación e identificación de

compuestos químicos en mezclas complejas.


22

La utilización de la GC-MS requiere sistemas especiales de conexión. En principio, se

trata de dos técnicas que trabajan en fase gaseosa y necesitan una muy pequeña cantidad

de muestra para su análisis, por lo que son muy compatibles. El único obstáculo serio a la

hora de realizar su acoplamiento es que el efluente que emerge de la columna

cromatográfica sale a presión atmosférica y debe introducirse en el interior del MS a alto

vacío. Actualmente, el acoplamiento directo resulta fácil cuando se utiliza la GC capilar, que

es el caso más habitual (Kitson et al., 1996; Santos et al., 2003).

Una mezcla de compuestos inyectada en el GC se separa en la columna

cromatográfica obteniendo la elución sucesiva de los componentes individuales aislados

que pasan inmediatamente al MS. Cada uno de estos componentes se registra en forma de

pico cromatográfico y se identifica mediante su respectivo espectro de masas. En este

proceso también se registra la corriente iónica total generada en la fuente iónica, cuya

representación gráfica constituye el cromatograma o “TIC” (cromatograma de iones

totales). En efecto, la corriente iónica generada por todos los iones da lugar a un pico

gaussiano de área proporcional a la concentración del compuesto detectado.

1.6. Aplicación de GC-MS a estudios de screening de contaminantes orgánicos en

muestras ambientales y de salud pública.

Las diferentes enfoques analíticos que ofrece el acoplamiento GC-MS con

analizadores TOF, hacen de esta técnica una herramienta realmente idónea para el

screening medioambiental de contaminantes orgánicos (Hernández et al., 2007; Hernández

et al., 2009). GC-TOF MS es una excelente técnica para la identificación de contaminantes

orgánicos de interés en distintos tipos de muestras de interés ambiental y alimentario.

Mediante GC-TOF MS es posible desarrollar metodología analítica avanzada para la rápida

detección y confirmación inequívoca de diferentes contaminantes. Esta metodología

permite el screening rápido y fiable de un amplio rango de contaminantes con poca

manipulación de muestra y con bajo consumo de disolventes.


23

En nuestro caso, la utilización de GC-TOF MS para el screening de contaminantes

orgánicos se llevó a cabo en muestras de sal, salmueras y aguas medioambientales.

Mediante esta técnica ha sido posible identificar compuestos de creciente interés

toxicológico. Entre ellos destacamos los alquilfosfatos y derivados (Reemtsma et al., 2008),

compuestos farmacéuticos o de cuidado personal como el Galoxolide (Muñoz et al., 2008)

así como compuestos filtrantes de la radiación UV como la Benzofenona (Jeon et al., 2006).

1.7. Aplicación de GC-MS a estudios de bioacumulación.

La bioacumulación es un proceso por el que un compuesto químico entra en un

organismo acuático directamente desde el agua a través de las agallas, del tejido epitelial o

directamente a través de la comida. Después de la entrada del compuesto químico en el

organismo, éste se distribuye gracias al sistema circulatorio por todos los tejidos, tendiendo

a depositarse preferentemente en órganos y tejidos grasos. La bioacumulación de un

compuesto químico se cuantifica en función de sus factores de bioacumulación o

bioconcentración o en el caso de no contar con estos valores se estima en función de su

coeficiente de partición octanol-agua (KOW). El valor de KOW de una sustancia está

relacionado con su potencial de bioacumulación. Un valor de KOW bajo indica movilidad y

transporte de ese compuesto, una fácil metabolización y biodegradación por lo que no

tiende a bioacumularse. Por el contrario, un valor de KOW alto indica posible bioacumulación

y absorción. La movilidad y acumulación de un contaminante dependerán de sus

características y de la naturaleza de los compartimientos ambientales.

La técnica de GC-MS ha sido y sigue siendo en la actualidad ampliamente utilizada

para la determinación de POPs en muestras complejas a niveles de traza gracias a su

excelente sensibilidad y selectividad. Para el estudio de las concentraciones de

contaminantes orgánicos hidrófobos, el potencial de la técnica GC-MS presenta grandes

aportaciones para su determinación proporcionando resultados satisfactorios (Serrano et

al., 2008; Van Leeuwen et al., 2008). GC-MS es una técnica ampliamente reconocida para el

análisis de compuestos bioacumulables dado su excelente potencial de separación,


24

identificación y cuantificación (Serrano et al., 2003 y 2008; Dórea et al., 2006; Hites et al.,

2004).

Los PBDEs, PAHs y OCs, incluyendo los PCBs, son contaminantes orgánicos que

muestran un carácter no polar, son altamente lipofílicos y presentan gran afinidad de

bioacumulación en tejidos de organismos vivos (Chiu et al., 2000; Braune et al., 2005). El

comportamiento bioacumulador se basa también en las funciones de los órganos

excretores (riñón, pulmón, etc.) que en muchos casos, no son capaces de activar rutas

detoxificadoras para reducir la presencia de estos contaminantes.

Las metodologías analíticas desarrolladas en esta Tesis se aplican al estudio de la

posible bioacumulación de PAHs, OCs y PBDEs en los productos de la acuicultura marina,

mediante su aplicación al análisis de piensos y filetes de pescado procedentes de un estudio

experimental que incluye un ciclo completo de engorde de doradas (Sparus aurata L.) en el

marco del Proyecto AQUAMAX (Sustainable Aquafeeds to Maximise the Health Benefits of

Farmed Fish for Consumers. www.aquamaxip.eu). Las muestras consideradas incluyen

materias primas de origen vegetal y animal, piensos con diferente composición y filetes de

dorada cultivada. El objetivo básico de este proyecto ha sido establecer el nivel máximo de

sustitución conjunta de proteínas y aceites de pescado en piensos de acuicultura

elaborados con materias primas alternativas, sostenibles y con una baja carga de

contaminantes, sin afectar al crecimiento, la conversión del alimento, la salud y bienestar

animal. Todo ello sin perjuicio de obtener un pescado de alto valor nutritivo, garantizando

la seguridad y calidad alimentaria, y con amplia aceptación por parte del consumidor.

1.8. Referencias.

Barker, J. 1998. Mass spectrometry: Analytical chemistry by open learning. New York, John

Wiley & Sons.

Björklund, E.; Sporring, S.; Wiberg, K.; Haglund, P.; Holst, C.V. 2006. New strategies for

extraction and clean-up of persistent organic pollutants from food and feed samples

using selective pressurized liquid extraction. Trends Anal. Chem. 25, 318-325.


25

Braune, B.M., Outridge, P.M., Fisk, A.T., Muir, D.C.G., Helm, P.A., Hobbs, K., Hoekstra, P.F.,

(...), Stirling, I. 2005. Persistent organic pollutants and mercury in marine biota of

the Canadian Arctic: An overview of spatial and temporal trends. Sci. Tot. Environ.

351-352, 4-56.

Cela, R.; Lorenzo, R.A.; Casais, M.C. 2002. Técnicas Analíticas de Separación en Química

analítica. Ed. Síntesis.

Chhabil, D. 2007. Fundamentals of contemporary mass spectrometry. University of

Memphis, Memphis, Tennessee, USA.

Chiu, A.; Chiu, N.; Beaubier, N.T.; Beaubier, J.; Nalesnik, R.; Singh, D.; Hill, W.R.; Lau, C.;

Riebow, J. 2000. Effects and mechanisms of PCB ecotoxicity in food chains: Algae →

fish → seal → polar bear. J Environ. Sci. Heal. C 18, 127-152.

Clement, R.E.; Yang, P.W.; Koester, C.J. 1999. Environmental analysis. Anal. Chem. 71, 257-

292.

Commission Decision 2002/657/EC implementing Council Directive 96/23/EC concerning

the performance of analytical methods and the interpretation of results. Available:

eur-lex. europa.eu/pri/en/oj/dat/2002/l_221/l_22120020817en000-80036.pdf.

Daughton C. G.; Ternes T. A. 1999. Pharmaceuticals and personal care products in the

environment: agents of subtle change. Environ. Health Perspect. 107, 907-938.

De Koning, S.; Janssen, H.-G.; Brinkman, U.A.Th. 2009. Modern methods of sample

preparation for GC analysis. Chromatographia 69, 33-78.

Document N° SANCO/3131/2007. Method validation and quality control procedures for

pesticide residues analysis in food and feed.

Document N° SANCO/10684/2009. Method validation and quality control procedures for

pesticide residues analysis in food and feed.


26

Dórea, J.G. 2006. Fish meal in animal feed and human exposure to persistent

bioaccumulative and toxic substances. J.Food Prot. 69, 2777-2785.

Eljarrat, E.; Barceló, D. 2003. Priority lists for persistent organic pollutants and emerging

contaminants based on their relative toxic potency in environmental samples. Trends

Anal. Chem. 22, 655-665.

Fernández-González, V.; Concha-Graña, E.; Muniategui-Lorenzo, S.; López-Mahía, P.; Prada-

Rodríguez, D. 2008a. Pressurized hot water extraction coupled to solid-phase

microextraction-gas chromatography-mass spectrometry for the analysis of polycyclic

aromatic hydrocarbons in sediments. J. Chromatogr. A 1196, 65-72.

Fernández-González, V.; Muniategui-Lorenzo, S.; López-Mahía, P.; Prada-Rodríguez, D.

2008b. Development of a programmed temperature vaporization-gas

chromatography-tandem mass spectrometry method for polycyclic aromatic

hydrocarbons analysis in biota samples at ultratrace levels. J Chromatogr. A 1207,

136-145.

Fidalgo-Used, N.; Blanco-González, E.; Sanz-Medel, A. 2007. Sample handling strategies for

the determination of persistent trace organic contaminants from biota samples. Anal.

Chim. Acta 590, 1-16.

Garrido Frenich, A.; Plaza-Bolaños, P.; Martínez Vidal, J.L. 2008. Comparison of tandem-in-

space and tandem-in-time mass spectrometry in gas chromatography determination

of pesticides: Application to simple and complex food samples. J. Chromatogr. A

1203, 229-238.

Griffitt, K.R.; Craun, J.C. 1974. Gel permeation chromatographic system: an evaluation. J.

Assoc. Off. Ana. Chem. 57, 168-172.

Hernández, F.; Sancho, J.V.; Pozo, O.J. 2005. Critical review of the application of liquid

chromatography/mass spectrometry to the determination of pesticide residues in

biological samples. Anal. Bioanal. Chem. 382, 934-946.


27

Hernández, F.; Portolés, T.; Pitarch, E.; Lopez, F.J. 2007. Target and non target screening of

organic micropollutants in water by solid-phase microextraction combined with gas

chromatography/high-resolution time-of-flight mass spectrometry. Anal. Chem. 79,

9494-9504.

Hernández, F.; Portolés, T.; Pitarch, E.; López, F.J. 2009. Searching for anthropogenic

contaminants in human breast adipose tissues using gas chromatography-time-of-

flight mass spectrometry. J. Mass Spectrom. 44, 1-11.

Hites, R.A.; Foran, J.A.; Carpenter, D.O.; Hamilton, M.C.; Knuth, B.A.; Schwager, S.J. 2004.

Global Assessment of Organic Contaminants in Farmed Salmon. Science 303, 226-

229.

Jeon, H.-K.; Chung, Y.; Ryu, J.-C. 2006. Simultaneous determination of benzophenone-type

UV filters in water and soil by gas chromatography-mass spectrometry. J.

Chromatogr. A 1131, 192-202.

Kitson F.G. 1996. Gas chromatography and mass spectrometry. A practical guide. San Diego,

Academic press.

McMaster, M.; McMaster, C. 1998. GC/MS. A practical user´s guide. United States of

America, Wiley-VCH.

Meloan C.E. 1999. Chemical Separations. Principles, Techniques and Experimentes. New

York, Ed. Wiley.

Muñoz, I.; Gómez, M. J.; Molina-Díaz, A.; Huijbregts, M. A. J.; Fernández-Alba, A. R.; García-

Calvo, E. 2008. Ranking potential impacts of priority and emerging pollutants in urban

wastewater through life cycle impact assessment. Chemosphere 74, 37-44.

Perugini, M.; Visciano, P.; Giammarino, A.; Manera, M.; Di Nardo, W.; Amorena, M. 2007.

Polycyclic aromatic hydrocarbons in marine organisms from the Adriatic Sea, Italy.

Chemosphere 66, 1904-1910.


28

Picó, Y.; Fernández, M.; Ruiz, M.J.; Font, G. 2007. Current trends in solid-phase-based

extraction techniques for the determination of pesticides in food and environment. J.

Biochem. Biophys. Methods 70, 117-131.

Pitarch, E.; Portolés, T.; Marín, J.M.; Ibáñez, M.; Albarrán, F.; Hernández, F. 2010. Analytical

strategy based on the use of liquid chromatography and gas chromatography with

triple-quadrupole and time-of-flight MS analyzers for investigating organic

contaminants in wastewater. Anal. Bioanal. Chem. 397, 2763-2776.

Poster, D.L.; Schantz, M.M.; Sander, L.C.; Wise, S.A. 2006. Analysis of polycyclic aromatic

hydrocarbons (PAHs) in environmental samples: A critical review of gas

chromatographic (GC) methods. Anal. Bioanal. Chem. 386, 859-881.

Reemtsma, T., García-López, M., Rodríguez, I., Quintana, J.B., Rodil, R. 2008.

Organophosphorus flame retardants and plasticizers in water and air I. Occurrence

and fate. Trends Anal. Chem. 27, 727-737.

Rubio, S.; Pérez-Bendito, D. 2009. Recent advances in environmental analysis. Anal. Chem.

81, 4601-4622.

Serrano, R.; Blanes, M.A.: López, F.J. 2008. Biomagnification of organochlorine pollutants in

farmed and wild gilthead sea bream (Sparus aurata L.) and stable isotope

characterization of the trophic chains. Sci. Total Environ. 389, 340-349.

Serrano, R.; Barreda, M.; Pitarch, E.; Hernández, F. 2003. Determination of low

concentrations of organochlorine pesticides and PCBs in fish feed and fish tissues

from aquaculture activities by gas chromatography with tandem mass spectrometry.

J. Sep. Sci. 26, 75-86.

Santos, F.J.; Galceran, M.T., 2003. Modern developments in gas chromatography-mass

spectrometry-based environmental analysis. J. Chromatogr. A 1000, 125-151.

Schantz, M.M. 2006. Pressurized liquid extraction in environmental analysis. Anal. Bioanal.

Chem. 386, 1043-1047.


29

Ternes, T.A.; Joss A; Siegrist, H. 2004.Scrutinizing pharmaceuticals and personal care

products in wastewater treatment. Environ. Sci. Technol. 38, 392-399.

Tindle, R.C.; Stalling, D.L. 1972. Apparatus for automated gel permeation cleanup for

pesticide residue analysis: Applications to fish lipids. Anal.Chem. 44, 1768-1773.

Van Leeuwen, S.P.J.; de Boer, J. 2008. Advances in the gas chromatographic determination

of persistent organic pollutants in the aquatic environment. J. Chromatogr. A 1186,

161-182.

CAPÍTULO 2

OBJETIVOS

Capítulo 2. Objetivos

33

El objetivo principal de la presente Tesis Doctoral es el desarrollo de metodología

analítica avanzada para la determinación de contaminantes orgánicos en muestras

medioambientales complejas.

La primera parte de la Tesis, dedicada al desarrollo de metodologías basadas en GC-

MS para la determinación de POPs en muestras marinas grasas, presenta los objetivos

parciales que se detallan a continuación:

1. Desarrollo de metodología analítica basada en GC-(QqQ)MS/MS y GC-TOF MS

para la determinación y confirmación de PAHs en matrices complejas.

2. Desarrollo de metodología analítica basada en el acoplamiento GC-MS para la

determinación de PBDEs en matrices complejas.

En la segunda parte, se desarrollan metodologías basadas en GC-TOF MS para el

análisis de la presencia de contaminantes orgánicos en muestras ambientales. Los objetivos

parciales son los siguientes:

1. Desarrollo de metodología “non-target” par la identificación de contaminantes

orgánicos en muestras de sal y agua de mar mediante GC-TOF MS.

2. Desarrollo y aplicación de metodología analítica avanzada para el estudio de OPEs

en sales, salmueras y aguas medioambientales mediante GC-TOF MS.

Por último, la aplicación de metodología analítica previamente desarrollada se aplicó

al análisis de muestras procedentes de un experimento de bioacumulación de

contaminantes orgánicos persistentes, lo que permitió el estudio de la posible acumulación

Capítulo 2. Objetivos

34

de estos en dorada a lo largo de un ciclo de crecimiento completo. Los objetivos parciales

son:

1. Estudio de la bioacumulación de PAHs en filetes de dorada.

2. Estudio de la bioacumulación de OCs en filetes de dorada.

CAPÍTULO 3

PLAN DE TRABAJO

Capítulo 3. Plan de trabajo

37

Para alcanzar los objetivos propuestos en esta Tesis se ha seguido el siguiente plan

de trabajo para el análisis cuantitativo:

• Selección de los contaminantes en función de su persistencia, toxicidad y carácter

emergente en muestras procedentes de la acuicultura.

• Revisión bibliográfica de procedimientos analíticos para la determinación de

contaminantes a niveles de traza en las matrices seleccionadas.

• Optimización de la separación cromatográfica mediante inyección en el equipo de

patrones de referencia. Selección del gradiente de temperaturas óptimo para los

compuestos seleccionados. Estudio del dwell time en función de la anchura de los

picos, del número de transiciones y la sensibilidad requerida.

• Estudio de las condiciones óptimas de MS mediante inyección de patrones de

referencia. Estudio de los espectros de ionización electrónica y selección de iones

precursores para cada analito. Para los PAHs, aislamiento de los iones precursores

seleccionados en el primer cuadrupolo y optimización de la energía de colisión para

la obtención de iones producto característicos, en el segundo cuadrupolo. Selección

de dos transiciones SRM por compuesto atendiendo a la sensibilidad y selectividad.

Para el caso de los PBDEs, selección de los iones más sensibles para su adquisición

en modo SIR (NCI-MS).

• Estudio y optimización de procedimientos de extracción y purificación para la

determinación de los contaminantes orgánicos seleccionados a niveles de traza en

las matrices seleccionadas.


38

• Estudio del efecto matriz en las muestras mediante el análisis de muestras

fortificadas y patrones en solvente a la misma concentración.

• Validación de metodología analítica para la determinación de PAHs y PBDEs en

muestras de peces, piensos y materias primas mediante el acoplamiento GC-MS

con analizador QqQ. Estudio de los parámetros de linealidad, especificidad,

exactitud, precisión y confirmación, mediante ensayos de recuperación en muestras

fortificadas a varios niveles de concentración atendiendo las guías SANCO de la

Unión Europea.

• Aplicación de la metodología validada para la determinación de PAHs y PBDEs en

muestras procedentes de la acuicultura, entre las cuales destacan los filetes de

dorada, piensos y materias primas, incluyendo aceites de pescado y aceites

vegetales.

En términos de análisis confirmativo mediante GC-TOF MS, se ha seguido el

siguiente plan de trabajo:

• Inyección de patrones y obtención del tiempo de retención de los fragmentos

iónicos más representativos con su masa exacta, a partir de su espectro en

ionización electrónica.

• Desarrollo de metodología de identificación mediante el software TargetLynx. Se

incluirá en la metodología información relevante para la identificación de cada

analito, como su tiempo de retención, composición elemental y masa exacta de los

principales fragmentos de su espectro de ionización electrónica, así como la

relación de intensidades entre los mismos.

• Inyección de las muestras en GC-TOF MS.

• Aplicación de la metodología analítica a muestras procedentes de la acuicultura.

Comparación de los datos obtenidos con los obtenidos previamente mediante GC-

(QqQ)MS/MS para la determinación de PAHs.


39

De manera general para el análisis cualitativo desarrollado en la presente Tesis

Doctoral, se ha seguido el siguiente plan de trabajo:

• Revisión bibliográfica y recopilación de información de contaminantes de interés así

como posibles contaminantes encontrados en muestras de agua de mar y sales

marinas.

• Desarrollo de metodología non-target mediante GC-TOF MS basada en el software

de deconvolución de datos ChromaLynx que proporciona, de manera automática, el

espectro de iones de masa exacta, permitiendo su comparación con los espectros

disponibles en librerías comerciales, para ionización electrónica, así como el estudio

del error de masas de los fragmentos seleccionados.

• Aplicación de tratamientos de muestra adecuados para la realización de rápidos

screening que incluyan una menor manipulación de la muestra, con el fin de

conseguir elevada multirresidualidad y que sea compatible con la identificación a

bajos niveles de concentración. Para ello se utilizará la técnica de SPE.

• Estudio de la presencia de contaminantes en muestras de sales marinas y agua de

mar mediante metodología non-target.

• Desarrollo de metodología target mediante GC-TOF MS, para compuestos éster

organofosforados (OPEs), mediante el uso del software de procesamiento de datos

TargetLynx. La metodología contendrá información relevante para la identificación

de cada analito, como su tiempo de retención, composición elemental y masa

exacta de los principales fragmentos de su espectro de ionización electrónica, así

como la relación de intensidades entre los mismos.

• Estudio de los parámetros críticos que pueden afectar al procesamiento de datos

target como la selección de los iones, ventana de extracción de masa,

interferencias con el lock mass, problemas de saturación del detector, entre otros.


40

• Aplicación de la metodología analítica de screening de OPEs desarrollada a

muestras de sales.

• Aplicación de la metodología analítica de screening de OPEs a muestras de sal,

salmueras y aguas medioambientales previamente analizadas mediante GC-TOF

MS. Análisis retrospectivo.

Por último, el plan de trabajo incluye la aplicación de metodologías analíticas

disponibles al estudio de la posible bioacumulación de OCPs, PAHs, PBDEs y PCBs en los

productos de la acuicultura marina, mediante un estudio experimental que incluye un ciclo

completo de engorde de doradas (Sparus aurata L.), y la interpretación de los resultados

obtenidos.

El presente trabajo ha quedado reflejado en seis artículos científicos, ya publicados y

pendientes de publicación, en revistas de carácter internacional.

CAPÍTULO 4

DETERMINACIÓN DE CONTAMINANTES ORGÁNICOS PERSISTENTES EN MUESTRAS

COMPLEJAS CON ALTO CONTENIDO GRASO.

Capítulo 4. Introducción

43

4.1. Introducción


44


45

El objetivo de este capítulo es el desarrollo de metodología analítica avanzada para la

determinación de POPs a niveles traza en matrices complejas procedentes de la acuicultura

marina. Las muestras consideradas incluyen piensos para el cultivo de peces, materias

primas para su elaboración (aceites vegetales y de pescado) y peces cultivados. Debido a la

complejidad de las muestras descritas y a la necesidad de alcanzar niveles de detección y

cuantificación lo más bajos posible, se ha prestado especial atención al tratamiento de

muestra, previo a la determinación instrumental mediante GC-MS.

En una primera parte se desarrolla metodología analítica para la determinación de

PAHs. Para ello se utilizó la técnica de GC-MS con analizadores QqQ para la cuantificación y

analizadores TOF para la confirmación adicional de los PAHs. La metodología desarrollada

resultó ser suficientemente rápida, sensible y selectiva para la determinación de PAHs en

muestras con alto contenido graso. Gracias a una eficiente extracción y purificación

mediante saponificación y posterior SPE, fue posible la validación de la metodología en

cuatro tipos de muestras complejas alcanzando LOQs a niveles bajos de concentración

(µg/Kg). La combinación de analizadores QqQ y TOF resultó una excelente estrategia para la

confirmación de la presencia de los PAHs en las muestras analizadas.

En la segunda parte del capítulo se desarrolla metodología para la determinación

de PBDEs mediante GC-MS en modo NCI. Esta metodología permitió la cuantificación y

confirmación de la identidad de 12 PBDEs en muestras con alto contenido graso. Como

resultado del tratamiento de muestra mediante digestión con ácido sulfúrico y posterior

SPE con Florisil, fue posible llevar a cabo la validación de la metodología en cuatro tipos de

muestras complejas alcanzando LOQs a niveles bajos de concentración (µg/Kg). Cabe

destacar que fue posible la determinación del BDE 209, uno de los retardantes de llama

más complicados de determinar mediante GC-MS por requerir condiciones cromatográficas

específicas. Trabajando con el analizador QqQ en modo SIR fue posible la adquisición de

tres iones carácterísticos por compuesto lo que permitió una correcta identificación de los

12 PBDEs seleccionados.

Capítulo 4. Determinación de PAHs mediante GC-(QqQ)MS/MS y GC-TOF MS

47

4.2. Desarrollo de metodología analítica para la determinación de hidrocarburos

policíclicos aromáticos mediante GC-(QqQ)MS/MS y GC-TOF MS.


48


49

4.2.1. Introducción.

Los PAHs son derivados poliméricos del benceno, causantes de reacciones adversas

no deseadas en animales, en humanos y en el medio ambiente. Los PAHs se forman a partir

de materia orgánica cuando ésta se somete a temperaturas elevadas durante un cierto

tiempo. Fuentes naturales como las erupciones volcánicas o los incendios forestales

también emiten PAHs al medio (Srogi et al., 2007). Otras fuentes de PAHs son las emisiones

de gases procedentes de la combustión de petróleo y carbón, así como de calefacciones. A

parte, el hábito de fumar, el consumo de alimentos ahumados o demasiado tostados por el

efecto del fuego puede incrementar la exposición del hombre a los PAHs hasta superar

niveles de riesgo. Los PAHs son compuestos ubicuos, por lo que la población está expuesta

a ellos por diferentes vías. Como consecuencia, es necesaria una legislación adecuada que

regule los niveles de estos contaminantes en los alimentos, con el fin de salvaguardar la

salud pública. En este sentido, el reglamento de la Unión Europea Nº 1881/2006 de la

comisión de 19 de diciembre de 2006, fija el contenido máximo de determinados

contaminantes en los productos alimenticios. La toxicidad del contenido de PAHs en una

muestra está referenciado al benzo(a)pireno. Así, se establecen contenidos máximos

referidos a concentración en peso fresco de la muestra (µg/Kg), como por ejemplo, para

filetes de pescado no ahumados, aceites y grasas, de 2 µg de benzo(a)pireno por Kg de

muestra.

Los PAHs constituyen una familia de compuestos ampliamente distribuida en el

medio ambiente, que se caracterizan por contener dos o más anillos de benceno unidos

entre sí. En la Figura 4.1. se muestran 16 PAHs considerados de riesgo y altamente tóxicos

por la Agencia de Protección Ambiental de los Estados Unidos (USEPA, www.epa.gov).

Los PAHs son compuestos lipófilos, con tendencia a acumularse en los tejidos grasos.

La toxicidad de los PAHs depende en gran medida de su estructura. Entre los más peligrosos

destacan los derivados del antraceno, así como las moléculas derivadas a las que se les

añade algún anillo bencénico, como en el caso del benzo[a]pireno y el

dibenzo[a,h]antraceno.


50

Figura 4.1. Estructuras de los 16 PAHs considerados cancerígenos por la USEPA.

La exposición humana a estos contaminantes se debe principalmente a la inhalación

o ingestión, aunque también es posible la intoxicación vía cutánea. Tras la absorción se

produce la distribución de estos hacia los órganos y tejidos, especialmente los de contenido

graso. Una vez incorporados al organismo, experimentan una oxidación enzimática llevada

a cabo en el hígado que los convierte en epóxidos y en dihidrodioles. Estas especies

químicas constituyen la forma genotóxica activa de los PAHs, las cuales pueden formar

aductos covalentes con proteínas y ácidos nucleicos celulares (Koss et al., 1999). Los

aductos con ADN generan mutación genética, de graves consecuencias por la generación de

tumores malignos para el individuo expuesto, sin olvidar la posibilidad de malformaciones

en embriones.

Los alimentos constituyen una vía de exposición a los PAHs para la población. Los

consumidores pueden estar expuestos a PAHs mediante la ingesta de alimentos como

cereales contaminados, harina, pan y verduras o alimentos braseados. Algunos autores han

encontrado PAHs a niveles de µg/kg en muestras de pescado y marisco comercializado en

mercados (Srogi et al., 2007). También, algunos cereales que han sufrido un proceso de

tostado, pueden presentar niveles elevados al igual que vegetales como el té que puede

contener un total de PAHs a niveles de µg/Kg (Lin et al., 2005). Entre los PAHs detectados,

Naphthalene Acenaphthylene Acenaphthene Fluorene Phenanthrene Anthracene

Fluoranthene Pyrene B(a) anthracene Chrysene B(b) fluoranthene B(k) fluoranthene

B(a) pyrene Indeno (1,2,3-c,d) pyrene Dibenzo(a,h) anthracene B(g,h,i) perylene


51

el de mayor interés es el benzo(a)pireno, ya que se ha determinado su toxicidad y, como se

ha comentado anteriormente, es el compuesto utilizado como referencia para establecer

límites legales en alimentos. En general, entre los alimentos no ahumados, los de mayor

contenido graso son los que presentan niveles superiores de benzo(a)pireno (Bartle et al.,

1991).

Como consecuencia de sus propiedades físico-químicas, los PAHs son compuestos

persistentes en el medio marino acumulándose en la fracción lipídica de los tejidos grasos

de los organismos marinos, dañando potencialmente los recursos marinos. Los mecanismos

de acumulación de estos compuestos en los organismos acuáticos pueden ser muy

diversos: por difusión pasiva del agua a través de las agallas o la piel (bioconcentración), a

través de la ingestión de partículas suspendidas contaminadas o, en mayor proporción, por

ingesta de alimento contaminado (procesos de bioacumulación y biomagnificación)(Van der

Oost et al., 2003). La bioacumulación está influenciada por numerosos procesos biológicos

como son la respiración, la capacidad de los organismos tanto de acumular como de

eliminar los compuestos y del crecimiento de la especie.

En la actualidad la acuicultura marina ha sufrido una importante expansión debido a

la sobreexplotación de los mares que ha supuesto la pesca extractiva y la creciente

demanda de pescado en el mercado. De hecho, el esfuerzo pesquero ha alcanzado su techo

y es urgente la búsqueda de fuentes alternativas de materias primas para los piensos de

engorde de peces. Actualmente, la industria pesquera vive días difíciles puesto que no

puede abastecer a la demanda global de pescado. Hasta ahora, la acuicultura ha logrado

cubrir este déficit pero su crecimiento se hace cada vez más complicado debido a la

necesidad de harinas y aceite de pescado, materias primas cada vez más limitadas y de las

cuales depende. Dentro de la acuicultura, hay también una creciente preocupación por la

presencia de compuestos tóxicos, tanto en materias primas como en los productos

procedentes de la acuicultura. (Aquamax newsletter, 2008).

Como respuesta al interés que presenta el análisis de estos contaminantes en peces,

piensos y materias primas utilizadas en la elaboración de estos piensos, se ha desarrollado

una metodología analítica que permita la determinación de PAHs a niveles traza en


52

diferentes matrices procedentes de la acuicultura marina, bajo unos criterios estrictos de

cuantificación y confirmación de la identidad de los mismos.

En el desarrollo de la metodología analítica se ha prestado especial atención a la

etapa de extracción y purificación debido al elevado contenido lipídico de las muestras

analizadas y la complejidad de las mismas.

En la etapa de la determinación instrumental se ha optimizado la cromatografía y la

adquisición mediante dos analizadores de MS. En primer lugar se ha utilizado un analizador

QqQ trabajando en modo MS/MS para la cuantificación a niveles de traza y en segundo

lugar se ha utilizado un analizador TOF para la confirmación inequívoca de la identidad de

los compuestos determinados con el analizador QqQ. La metodología desarrollada se ha

aplicado para la determinación de PAHs en muestras reales procedentes de la acuicultura

marina de manera fiable y precisa tanto para la cuantificación como para la identificación

inequívoca de los mismos.

Las matrices incluidas son filetes de dorada cultivada, piensos con diferentes

proporciones de aceites de origen animal y vegetal, aceite de pescado, aceite de lino, aceite

de colza y aceite de palma. Las muestras proceden de un estudio experimental de

crecimiento de doradas (Sparus aurataL.) en el marco del Proyecto AQUAMAX

(www.aquamaxip.eu).


53

4.2.2. Artículo científico 1:

A reliable analytical approach based on gas chromatography coupled to triple

quadrupole and time of flight analyzers for the determination and confirmation of

polycyclic aromatic hydrocarbons in complex matrices from aquaculture activities.

J. Nácher-Mestre, R. Serrano, T. Portolés, F. Hernández, L. Benedito-Palos, J. Pérez-

Sanchez



54


55

The potential of gas chromatography coupled to tandem mass spectrometry

(GC/MS/MS) with a triple quadrupole analyzer (QqQ) has been investigated for

the quantification and reliable identification of sixteen polycyclic aromatic

hydrocarbons (PAHs) from the EPA priority list in animal and vegetable

samples from aquaculture activities, whose fat content ranged from 5 to 100%.

Matrices analyzed included fish fillet, fish feed, fish oil and linseed oil.

Combining optimized saponification and solid-phase extraction led to high

efficiency in the elimination of interfering compounds, mainly fat, from the

extracts. The developed procedure minimized the presence of these

interfering compounds in the extracts and provided satisfactory recoveries of

PAHs. The excellent sensitivity and selectivity of GC/(QqQ)MS/MS in selected

reaction monitoring (SRM) allowed to reach limits of detection at pg/g levels.

Two SRM transitions were acquired for each analyte to ensure reliable

identification of compounds detected in samples. Confirmation of positive

findings was performed by GC coupled to high-resolution time-of-flight mass

spectrometry (GC/TOFMS). The accurate mass information provided by

GC/TOFMS in full acquisition mode together with its high mass resolution

makes it a powerful analytical tool for the unequivocal confirmation of PAHs in

the matrices tested. The method developed was applied to the analysis of real-


56

world samples of each matrix studied with the result of detecting and

confirming the majority of analytes at the µg/kg level by both QqQ and TOF

mass spectrometers.

INTRODUCTION

Polycyclic aromatic hydrocarbons (PAHs) are widespread environmental pollutants

from both natural and anthropogenic origins, such as the partial combustion of organic

compounds, or pollution from petrochemical activities.1 PAHs are lipophilic contaminants

and tend to accumulate in the biotic compartment of the environment.1-6

Concern has

arisen about their potential adverse effects on organisms, including human beings, and this

has led to the inclusion of 16 PAHs in the list of priority contaminants by the United States

Environmental Protection Agency.7

Marine aquaculture has undergone strong development in the last few decades as a

consequence of increased fish consumption by the world population and decreasing wild

stocks. Fish culture operates in parallel to traditional fisheries and nowadays both cultured

and wild fish are important components of the Mediterranean diet.8,9

Aquaculture products

are subject to increasingly strict control and regulation. As an example, the European

Commission Regulation (EC) 1881/200610

has fixed a maximum level of 2 µg/kg (wet

weight) for benzo[a]pyrene in fish.

PAHs enter the marine environment through atmospheric depositions and surface

runoff, and, due to their lipophilic character, they are accumulated by marine organisms.2-

4,6,11-13 As consequence of the artificial food chain in aquaculture activities, fish used as raw

material for the manufacture of fish feed ingredients (fish oils and meals) are a potential

source of PAHs in fish feed, which might be bioaccumulated by cultured fish.14

Likewise,

vegetable oils used in fish feed manufacture are another possible source of this family of

contaminants.5,15

The determination of PAHs in aquaculture matrices is difficult due to their complexity

and the presence of interfering substances, mainly fats that are co-extracted with the

analytes when using techniques such as Soxhlet, microwave-assisted extraction,


57

pressurized liquid extraction, etc.6,16-22

making an efficient cleanup before analysis essential.

Traditionally, the cleanup step is performed by solid-phase extraction (SPE) or preparative

column chromatography using Florisil, silica, alumina or other available absorbents.6,18,21,23

Other techniques are now applied, such as accelerated solvent extraction followed by gel-

permeation chromatography,19

solid-phase microextraction,22

or microwave-assisted

extraction.21

High-performance liquid chromatography (HPLC) with fluorescence detection has

been widely used for determination of PAHs.16

However, in the analysis of complex

matrices, gas chromatography (GC) rather than LC is often preferred for separation,

identification and quantification because GC generally affords greater selectivity, resolution

and sensitivity for determination of PAHs. GC/single quadrupole mass spectrometry has

been widely used for the determination of organic compounds in environmental

samples.13,15,19

In recent years, tandem mass spectrometry (MS/MS) has been increasingly

used because of its higher sensitivity and selectivity that minimize or even remove many

interferences. GC/MS/MS, using ion trap or triple quadrupole (QqQ) analyzers, has been

successfully applied to the analysis of PAHs in a variety of matrices.21,24

The use of two

stages of mass analysis in MS/MS systems based on QqQ analyzers offers the possibility of

applying selected reaction monitoring (SRM), one of the most selective and sensitive

approaches for quantification and confirmation, especially at trace levels. Thus, the number

of GC/MS/MS (QqQ) applications in the environmental and food analysis fields has

increased in the last few years.

In recent years, the use of high-resolution time-of-flight mass spectrometry (TOFMS)

has become prevalent in environmental analysis. This technique can provide conclusive

information for the reliable confirmation of target analytes and also for the structural

elucidation of non-target compounds due to its unrivalled full spectrum sensitivity together

with its greater mass resolution and mass accuracy.25,26

The aim of this study is to develop a reliable and sensitive methodology based on the

use of advanced GC/MS techniques for the quantitative determination and safe

identification of the 16 EPA PAH priority contaminants in complex matrices from

aquaculture activities. We have studied the extraction and cleanup steps, in order to obtain


58

extracts with minimal fat content. The developed methodology has been applied to real-

world samples of fish feed, raw materials and fish specimens from feeding trials performed

with gilthead sea bream (Sparus aurata L.) carried out as a part of the European Union

project Sustainable Aquafeeds to Maximise the Health Benefits of Farmed Fish for

Consumers (AQUAMAX; Contract number: 016249-2). The acquisition of two selective SRM

transitions for each target analyte by GC/(QqQ)MS/MS has allowed the quantification and

identification of PAHs at the low µg/kg level. The positive GC/(QqQ)MS/MS findings have

been confirmed by GC/TOFMS, taking advantage of the high mass resolution and mass

accuracy provided by this technique.

EXPERIMENTAL

Materials and reagents

A PAH analytical standard mixture (PAH-Mix 9) containing naphthalene,

acenaphthylene, acenaphthene, fluorene, phenanthrene, anthracene, fluoranthene,

pyrene, benzo[a]anthracene, chrysene, benzo[b]fluoranthene, benzo[k]fluoranthene,

benzo[a]pyrene, indeno[1,2,3-cd]pyrene, dibenzo[a,h]anthracene and benzo[g,h,i]perylene

was purchased from Dr Ehrenstorfer (Promochem, Wesel, Germany) with a purity 97-99.8%

as a 10 µg/mL solution in cyclohexane. Stock solutions (1 µg/mL) were prepared by

dissolving the reference standard in n-hexane and stored in a freezer at -20°C. Working

solutions were prepared by diluting the stock solution in n-hexane for sample fortification

and calibration curves.

In addition, benzo[a]anthracene-D12 from Dr Ehrenstorfer was used as a surrogate

internal standard in the validation study and real-world samples analysis. Working solutions

of the labeled standards were prepared by dilution of the commercial solutions with n-

hexane and stored at 4°C.

Methanol, dichloromethane, n-hexane and ethyl acetate (ultratrace quality) were

purchased from Scharlab (Barcelona, Spain). Anhydrous sodium sulfate of pesticide residue

quality (Scharlab) was dried for 18 h at 300°C before use. Potassium and sodium hydroxide

were purchased from Scharlab. The 0.2 µm filters were from Serviquimia (Castellón, Spain).


59

Silica cartridges (Strata 0.5 g and 1 g; Phenomenex, Torrance, CA, USA) and Supelclean LC-

Florisil SPE tubes (0.5 g and 1 g; from Sigma-Aldrich, Madrid, Spain) were used in SPE

experiments.

Sample material

Gilthead sea bream (Sparus aurata L.) specimens of Atlantic origin (Ferme Marine de

Douhet, Ile d'Oléron, France) were cultured at the Instituto de Acuicultura de Torre la Sal,

Castellón, Spain (IATS, CSIC) and collected when the fish reached commercial size (500 g).

The left-side fillets (denuded from skin and bone) were excised and stored at -20°C until

analysis. Samples of the fish feed supplied to sea bream during feeding trials from

AQUAMAX project experiments were stored at -20°C until analysis. Fish oil and linseed oil

used in fish feed manufacture, usually as raw materials of fish feed, were also stored at -

20°C until analysis. The ingredients and the chemical composition of the fish feed are

shown in Table 1.

GC instrumentation

Two GC systems (Agilent 6890N; Agilent, Palo Alto, CA, USA) equipped with an

autosampler (Agilent 7683) were coupled to (1) a triple quadrupole mass spectrometer

(Quattro Micro GC; Micromass, Boston, MA, USA) and (2) a time-of-flight mass

spectrometer (GCT; Waters Corp., Manchester, UK), both operating in electron ionization

(EI) mode. In both cases, the GC separation was performed using a fused-silica HP-5MS

capillary column (length 30 m, i.d. 0.25 mm, film thickness 0.25 µm; J&W Scientific, Folson,

CA, USA). The injector temperature was set to 250°C. Splitless injections of 1 µL of the

sample were carried out. Helium (99.999%; Carburos Metálicos, Valencia, Spain) was used

as carrier gas at a flow rate of 1 mL/min. The interface and source temperatures were set to

250°C in both systems and a solvent delay of 3 min was selected.


60

Table 1. Ingredients and chemical composition of fish feed analyzed.

Ingredient (%) FO 33VO 66VO

Fish meal (CP 70%) 1 15 15 15

CPSP 90 2 5 5 5

Corn gluten 40 40 40

Soybean meal 14.3 14.3 14.3

Extruded wheat 4 4 4

Fish oil 3 15.15 10.15 5.15

Rapeseed oil 0 0.85 1.7

Linseed oil 0 2.9 5.8

Palm oil 0 1.25 2.5

Soya lecithin 1 1 1

Binder 1 1 1

Mineral premix 4 1 1 1

Vitamin premix 5 1 1 1

CaHPO4.2H2O (18%P) 2 2 2

L-Lys 0.55 0.55 0.55

Composition

Dry matter (DM, %) 93.13 92.9 92.77

Protein (% DM) 53.2 52.81 52.62

Fat (% DM) 21.09 21 20.99

Ash (% DM) 6.52 6.69 6.57

1Fish meal (Scandinavian LT)

2Fish soluble protein concentrate (Sopropêche, France)

3Fish oil (Sopropêche, France)

4Supplied the following (mg · kg diet

-1, except as noted):

calcium carbonate (40% Ca) 2.15 g, magnesium hydroxide

(60% Mg) 1.24 g, potassium chloride 0.9 g, ferric citrate 0.2 g,

potassium iodine 4 mg, sodium chloride 0.4 g, calcium

hydrogen phosphate 50 g, copper sulphate 0.3, zinc sulphate

40, cobalt sulphate 2, manganese sulphate 30, sodium

selenite 0.3. 5Supplied the following (mg · kg diet

-1): retinyl acetate 2.58,

DL-cholecalciferol 0.037, DL-α tocopheryl acetate 30,

menadione sodium bisulphite 2.5, thiamin 7.5, riboflavin 15,

pyridoxine 7.5, nicotinic acid 87.5, folic acid 2.5, calcium

pantothenate 2.5, vitamin B12 0.025, ascorbic acid 250,

inositol 500, biotin 1.25 and choline chloride 500.

FO: Reference fish feed; 33VO: fish feed with 33% fish oil

replacement; 66VO: fish feed with 66% fish oil replacement

The oven temperature program in the GC/QqQ system was as follows: 90°C (1 min);

10°C/min to 250°C; 5°C/min to 300°C (3min). The oven temperature in the GC/TOF analysis


61

was programmed as follows: 90°C (1min); 5°C/min to 300°C (2 min). In both cases helium

was used as the carrier gas at a flow rate of 1 mL/min. The QqQ system was operated in

MS/MS mode using 99.995% argon (Carburos Metálicos) as the collision gas at a pressure of

0.28 Pa in the collision cell, and a dwell time per channel of between 0.1 and 0.3 s. The TOF

mass spectrometer was operated at 1 spectrum/s, acquisition rate over the mass range m/z

50-300, using a multichannel plate voltage of 2650 V. The mass resolution of the TOFMS

instrument was approximately 7000 (FWHM). Heptacosa standard, used for the daily mass

calibration and as lock mass, was injected via a syringe in the reference reservoir at 30°C,

monitoring the ion at m/z 218.9856. The application managers TargetLynx and QuanLynx

(Waters) were used to process the qualitative and quantitative data obtained from

calibration standards and from sample analysis.

Analytical procedure

Before analysis, samples were thawed at room temperature and, in the case of fish

fillet and fish feed, were carefully ground using a Super JS mill (Moulinex, Ecully Cedex,

France). Approximately 2 g of sample were homogenized with 6 g of anhydrous sodium

sulfate and the blend was spiked with 100 µL of surrogate solution (25 ng/mL). The 10 mL

of methanolic 1 M KOH solution were added to the mixture and saponified for 3 h at 80°C.

The analytes were then extracted twice with 8 mL of n-hexane and the solution was filtered

through a 0.2 µm filter and concentrated under a gentle stream of nitrogen at 40°C to a

volume of 1 mL (in the case of oils, extracts were concentrated to 5 mL). The 1mL extract

was passed through a Florisil SPE cartridge, previously conditioned with 6 mL of n-hexane,

and eluted with 8 mL dichloromethane/hexane (DCM/Hx) (20:80). The eluate was

evaporated under a gentle stream of nitrogen at 40°C and the final residue was redissolved

in 0.25 mL of n-hexane. The final extracts obtained after cleanup were analyzed by

GC/QqQ-MS under the experimental conditions shown in Table 2. Quantification of the

samples was carried out by means of calibration curves with standard in solvent using the

internal standard method. Positive real samples were reanalyzed by GC/TOFMS for

additional confirmation of the compounds detected by QqQ-MS.


62

Table 2. Experimental conditions of the optimized GC-EI(SRM) method.

tR (min) Window

(min) Compounds

Precursor

Ion (m/z)

Product

Ion (m/z) Q/q

Dwell time

(sec)

Collision

Energy

(eV)

Q/qc

Ratio

5.84 3-7 Naphthalene 128 102 Q

0.1 30

1.20 (5) 128 77 q 30

8.8 7-9.1 Acenaphthylene 152 126 Q

0.2 20

1.09 (6) 152 150 q 30

9.2 8.9-10.5 Acenaphthene 154 152 Q

0.2 35

3.58 (5) 153 126 q 30

10.44 9.7-11.5 Fluorene 165 115 Q

0.15 30

1.03 (10) 166 164 q 35

12.67 11.5-14

Phenanthrene b 178 152 Q 0.1 20 3.12 (4)

12.8 Anthracene b

178 176 q 0.1 35 2.62 (9)

15.51 14-15.9 Fluoranthene 202 200 Q

0.1 35

3.41 (5) 202 150 q 45

16.02 15.8-17 Pyrene 202 200 Q

0.1 30

28.17 (3) 202 152 q 20

18.9

17-21

B(a)Anthracene-D12a 240 236 0.1 30

19.1 B(a)Anthracene b 228 226 Q 0.1 20 4.00 (11)

19.2 Chrysene b

228 224 q 0.1 55 3.48 (12)

22.33 21-23

B(b)Fluoranthene b 252 250 Q 0.2 35 4.38 (7)

22.42 B(k)Fluoranthene b 250 248 q 0.2 35 4.02 (10)

23.31 23-25 B(a)Pyrene 252 250 Q

0.2 35

4.24 (4) 250 248 q 30

26.75 25-27.6 Indeno(1,2,3-cd)Pyrene 276 274 Q 0.1 40 2.89 (10)

276 272 q 60

26.9 Dibenzo(a,h)Anthracene 278 276 Q 0.1 30 3.27 (9)

278 274 q 30

27.5 27.2-28.5 B(g,h,i)Perylene 276 274 Q 0.3 30 3.54 (7)

274 272 q 30 aInternal Standard used as surrogate.

bThe same transitions for both compounds.

cAverage value

calculated from standard solutions at eight concentration levels each injected three times and RSD,

in brackets. Q: Quantification transition, q: confirmation transition.

Validation

Statistical validation of the GC/(QqQ)MS/MS method was performed by evaluating

the following parameters.

–Linearity. The calibration curves were obtained by injecting reference standard

solutions in triplicate. The concentration range tested was 0.5-200 ng/mL (eight


63

points). Linearity was assumed when the regression coefficient was greater than 0.99

with residuals randomly distributed and less than 30%.

–Accuracy. The accuracy was evaluated by means of recovery experiments, analyzing

blank samples of each matrix spiked at three concentration levels in fish fillet and fish

feed (0.125, 1.25 and 2.5 µg/kg, in sextuplicate) and two levels (1.25 and 2.5 µg/kg) in

fish oil and linseed oil (in triplicate). Previously, blank samples were analyzed to

determine the concentration of the analytes present in the matrices (fish fillet and fish

feed in sextuplicate, and oils in triplicate) (see Tables 3 and 4).

–Precision. The precision, expressed as the repeatability of the method, was determined

in terms of relative standard deviation in percentage (RSD, %) from recovery

experiments at each fortification level.

–Limit of quantification (LOQ). The LOQ was established as the lowest concentration

that was validated following the overall analytical procedure with satisfactory recovery

(between 70-110%) and precision (RSD <20%).

–Limit of detection (LOD). The LOD was statistically estimated, from the quantification

transition, as the analyte concentration giving a peak signal of three times the

background noise from the chromatograms at the lowest fortification level tested.

When the analytes were present in the blank, LODs were calculated from the

chromatogram of the analyzed blank sample. The LOD was calculated using the

software option for estimating the signal-to-noise (S/N) ratio and referring this value

to an S/N value of 3.27

–Confirmation criteria. The Q/q ratio, defined as the ratio between the intensity of the

quantification (Q) and the confirmation (q) transitions, was used to confirm the

identity of the compounds detected in samples. A safe confirmation was assumed by

taking into account of the European Commission Decision (2002/657/CE).28

Briefly, to

confirm a finding as an actual positive, a maximum ratio tolerance ±20% was accepted

when the relative intensity of the confirmative transition was >50% of the quantitative

one (Q/q ratio 1-2). For higher Q/q ratios, the tolerances increased. Thus, deviations

±25% (relative intensity 20-50%, Q/q ratio 2-5), ±30% (relative intensity 10-20%, Q/q


64

Table 3. Validation parameters obtained for the analysis of PAHs in fish fillets and fish feeds (n=6, at each fortification level). Analysis performed

by GC-(QqQ)MS/MS.

Recoveries (%) (n=6)

Fortification levels (µg/kg)

“Blank” (µg/kg)

(n=6) 0.125 1.25 2.5

LOD (µg/kg, fresh

weight)

Compound Fish Fillet

Fish

Feed

Fish

Fillet

Fish

Feed

Fish

Fillet

Fish

Feed

Fish

Fillet

Fish

Feed Fish Fillet

Fish

Feed

Naphthalene 2.4 (6) 2.9 (8) -* -* 105(36) 76(36) 107(37) 83(37) 0.06 0.09

Acenaphthylene 0.2(16) 0.2(2) 112(10) 80(20) 85(16) 85(17) 80(12) 61(17) 0.05 0.09

Acenaphthene 0.2 (17) 0.3 (15) 106(20) 102(27) 73(13) 108(5) 64(9) 87(13) 0.02 0.1

Fluorene 0.8 (6) 1.7 (6) 108(22) 82(16) 91(12) 104(13) 89(10) 108(12) 0.1 0.1

Phenanthrene 2.6 (20) 3.9 (3) -* -* 71(9) 92(4) 71(9) 102(8) 0.1 0.1

Anthracene 0.6 (14) 0.8 (1) 90(6) 100(22) 88(7) 101(7) 91(4) 96(12) 0.03 0.1

Fluoranthene 1.7 (1) 4(4) -* -* 76(14) 124(5) 86(7) 105(10) 0.1 0.1

Pyrene 4.4 (1) 7(1) -* -* 84(18) 105(2) 89(4) 99(6) 0.1 0.1

Benzo(a)anthracene 0.06 (19) 0.7 (4) 88(17) 83(15) 103(2) 99(2) 104(4) 108(4) 0.02 0.1

Chrysene 0.1 (19) 1.5 (3) 83(7) 98(25) 101(4) 101(2) 100(2) 108(9) 0.04 0.1

Benzo(b)fluoranthene - 0.8 (5) 93(9) 93(19) 100(4) 100(5) 100(3) 108(8) 0.06 0.1

Benzo(k)fluoranthene - 0.3 (17) 93(13) 103(6) 99(4) 102(2) 101(3) 102(3) 0.09 0.1

Benzo(a)pyrene - 0.45 (6) 98(8) 95(9) 101(3) 100(6) 101(3) 103(5) 0.09 0.09

Indeno (1,2,3-cd)pyrene - 0.3 (6) 97(13) 97(13) 96(6) 98(6) 97(7) 98(9) 0.09 0.07

Dibenzo(a,h)anthracene - 0.2 (16) 98(12) 92(14) 94(11) 81(3) 93(8) 74(9) 0.09 0.06

Benzo(g,h,i)perylene - 0.4 (8) 105(10) 109(9) 99(4) 101(4) 100(3) 104(10) 0.07 0.1

*- not validated due to the high analyte content in the “blank” sample.


65

Table 4. Validation parameters obtained for the analysis of PAHs in fish oil and linseed oil (n=3, at each fortification level). Analysis performed

by GC-(QqQ)MS/MS.

Recoveries (%) (n=3)

Fortification levels (µg/kg) LOD (µg/kg, lipid

weight) “Blank” (µg/kg) (n=3) 1.25 2.5

Compound Fish Oil

Linseed

Oil Fish Oil

Linseed

Oil Fish Oil

Linseed

Oil Fish Oil

Linseed

Oil

Naphthalene - - 78(28) 75(24) 80(21) 71(22) 0.4 0.1

Acenaphthylene - - 116(1) 74(21) 89(29) 82(24) 1 0.9

Acenaphthene 0.3 (8) - 97(8) 88(15) 105(28) 90(13) 0.2 1.25

Fluorene 4.3 (5) 2.6(4) 112(3) 87(13) 100(25) 115(4) 0.8 1.25

Phenanthrene 38.2 (2) 15.2(1) -* -* 113(9) 93(3) 1.25 1.25

Anthracene - - 94(11) 71(4) 104(8) 93(7) 1.25 1.25

Fluoranthene 13.3(7) 16.7(4) -* -* 98(3) 105(1) 1.25 1.25

Pyrene 9(1) 11.3(1) -* -* 92(1) 111(2) 1.25 1.25

Benzo(a)anthracene - - 113(7) 104(15) 112(4) 100(1) 0.6 1.25

Chrysene - - 119(3) 108(9) 104(5) 111(4) 0.6 1.25

Benzo(b)fluoranthene 0.3(22) 0.5(4) 111(6) 112(6) 111(4) 120(4) 0.3 0.4

Benzo(k)fluoranthene 0.3(2) 0.4(5) 108(6) 115(8) 107(4) 103(9) 0.3 0.4

Benzo(a)pyrene - 0.3(8) 80(5) 87(11) 110(4) 95(4) 0.7 0.3

Indeno (1,2,3-cd)pyrene - - 84(10) 107(14) 97(3) 103(7) 0.5 0.4

Dibenzo(a,h)anthracene - - 89(7) 99(19) 98(7) 117(2) 0.15 0.2

*- not validated due to the high analyte content in the “blank” sample.


66

ratio 5-10) and 50% (relative intensity 10%, Q/q ratio >10) were accepted. This criterion was

originally defined on measures to monitor certain substances and residues thereof in live

animals and animal products, and it is being increasingly used in other fields such as

environmental and biological samples analysis.29,30

Obviously, agreement in the retention

time for sample and reference standard was also required to confirm a positive finding. The

Q/q ratio for each compound was empirically determined as the average value calculated

from eight standard solutions injected in triplicate.

RESULTS AND DISCUSSION

Cleanup optimization.

Cleanup is an important step in environmental and food analysis due to the

complexity of the matrices and the selectivity required. Traditionally, analytical methods for

PAHs in fatty samples include an alkaline saponification. This treatment provides a

satisfactory lipid removal of the extracts.5,6,31

In our work, as a consequence of the

complexity of the matrices analyzed (especially fish and vegetable oils), modification of the

commonly applied saponification procedures was necessary in order to obtain complete

reaction of the lipids.6,32

Different alkaline solutions were investigated, including

NaOH/MeOH 1M, NaOH/EtOH 1M, NaOH/MeOH saturated, NaOH/EtOH saturated and

KOH/MeOH 1M, KOH /EtOH 1M, KOH/MeOH saturated, and KOH/EtOH saturated. The

effect of adding hexane or water was also tested but poor results were observed, so their

addition was discarded. In all cases, heating for more than 2 h at a temperature higher than

60°C was required to remove lipid interferences and to hydrolyze the lipids and esters

produced in the process. As can be seen in Fig. 1, fluorene, fluoranthene and pyrene were

masked when the saponification time was less than 3 h. Under those conditions, esters and

free fatty acids were not hydrolyzed, thus producing matrix effects. On the contrary, when

the sample was submitted to saponification for 3 h or more at 80°C, the matrix effect

decreased, and the selectivity and sensitivity improved. Thus, poor saponification led to

unsatisfactory data. These results are in agreement with previous data reported for low

recoveries of organic compounds in marine matrices, suggesting poor saponification as the


67

cause.6,31

We obtained the most satisfactory results when using KOH/MeOH 1 M for 3 h at

80°C (Fig. 1). Under these conditions, the saponification process removed most of the fats

after extraction with n-hexane and filtration (see analytical procedure).

Figure 1. GC/TOFMS extracted ion chromatograms (mass window 0.02 m/z units) for fluorene (m/z

166.0782), and fluoranthene and pyrene (m/z 202.0782) in fortified fish fillet extracts (50 µg/kg)

submitted to saponification at 80°C for (A) 1 h and (B) 3 h.

An additional step using SPE cleanup was still necessary to remove several interfering

compounds. Silica and Florisil cartridges with 0.5 g and 1 g stationary phase, and different

elution procedures, were compared with regard to their efficiency in lipid removal, the

recovery of analytes, elution volume and elution time.


68

Florisil cartridges (1 g) were finally selected, and the elution was performed with

DCM/Hx (20:80), which led to cleaner extracts and better recoveries than other mixtures

assayed, such as n-hexane, DCM or ethyl acetate. Using DCM/Hx (20:80), interfering

compounds seemed to be more retained on the cartridge, while PAHs eluted in cleaner

eluates.

Other SPE purification procedures reported, using silica, Florisil18

or C18 cartridges,33

did not allow the determination of light PAHs such as naphthalene and acenaphthylene,

this being a notable difference with the methodology developed in the present work.

GC/MS/MS optimization

High-resolution GC coupled to MS using either a single quadrupole or an ion trap for

detection has been widely used in environmental applications.16,19,34

However, there is little

information about the use of tandem mass spectrometry with QqQ for the determination

of PAHs in complex aquaculture samples. Optimization of the GC/MS/MS method was

performed in this study by injecting hexane standard solutions into the GC/(QqQ)MS/MS

system operating in EI mode. Full-scan EI mass spectra for all the PAH congeners showed

basically the molecular ion, with poor fragmentation. Thus, the molecular ion was selected

as the precursor ion for fragmentation purposes in the collision cell. Different values of

collision energy (between 20 and 60 eV) were tested to perform the subsequent

fragmentation of the selected precursor ion. Two MS/MS transitions were selected for each

compound, normally the most sensitive ones, in order to have a reliable confirmation of the

identity of the analyte. In most cases, the losses from the molecular ion of two or four

hydrogen atoms were chosen as the quantitative and/or confirmative transitions for the

determination of PAHs with improved selectivity and sensitivity. The dwell time parameter

was also optimized between 0.1 and 0.3 s in order to obtain at least ten points across the

peaks and thus maintain satisfactory sensitivity for each compound. Table 2 shows the

precursor and product ions corresponding to the quantitative and confirmation transitions

monitored. Linearity of the relative response of analytes was established by analyzing

hexane standard solutions in the ranges 0.5-200 ng/mL. Regression coefficients above


69

0.995 were obtained for all the compounds with residuals lower than 30% and without any

clear trend in their distribution.

GC/TOF MS optimization

In our study, GC/TOFMS was used for additional confirmation of the PAHs detected

in samples by QqQ-MS. For these purpose a TargetLynx processing method was employed,

using reference standard solutions in solvent. The MS spectrum for each compound was

obtained and four ions were selected, for which elemental compositions were proposed

(see Table 5). Narrow mass windows of 0.02 m/z units were chosen as a compromise

between sensitivity, peak shape and accurate mass measurements. Q/q intensity ratios

Table 5. Experimental ions selected for the confirmation of PAHs by GC-TOF MS.

Compound Ion 1 m/z 1 Ion 2 m/z 2 Ion 3 m/z 3 Ion 4 m/z 4

Naphthalene C10H8 128.0626 C10H7 127.0548 C10H6 126.0452 C8H6 102.047

Acenaphthylene C12H8 152.0626 C12H7 151.0548 C12H6 150.047 C10H6 126.047

Acenaphthene C12H9 153.0707 C12H10 154.0782 C12H8 152.0626 C10H6 126.047

Fluorene C13H9 165.0704 C13H10 166.0782 C13H8 164.0621 C11H7 139.0544

Phenanthrene C14H10 178.0782 C14H8 176.0626 C12H8 152.0626 C12H6 150.047

Anthracene C14H10 178.0774 C14H8 176.0626 C12H8 152.0626 C12H6 150.047

Fluoranthene C16H10 202.0782 C16H8 200.0626 C14H6 174.047 C12H6 150.047

Pyrene C16H10 202.0782 C16H9 201.0621 C16H8 200.0621 C14H6 174.047

Benzo(a)anthracene C18H12 228.0939 C18H10 226.0783 C16H8 202.0626 C9H6 114.047

Chrysene C18H12 228.0939 C18H10 226.0783 C16H8 202.0626 C9H6 114.047

Benzo(b)fluoranthene C20H12 252.0939 C20H10 250.0783 C10H6 126.047 C9H5 113.0391

Benzo(k)fluoranthene C20H12 252.0939 C20H10 250.0783 C10H6 126.047 C9H5 113.0391

Benzo(a)pyrene C20H12 252.0939 C20H10 250.0783 C10H6 126.047 C9H5 113.0391

Indeno (1,2,3-cd)pyrene C22H12 276.0939 C22H10 274.0783 C11H6 138.047 C10H5 125.0391

Dibenzo(a,h)anthracene C22H12 276.0939 C22H10 274.0783 C11H7 139.0548 C10H5 125.0391

Benzo(g,h,i)perylene C22H12 276.0939 C22H10 274.0783 C11H6 138.047 C10H5 125.0391

were used as confirmation parameter. Theoretical Q/q ratios were calculated from solvent

standard solutions as the ratio between the most sensitive ion (Q, quantitative) and each of

the other measured ions (q, confirmative). The selection of four ions provided up to three

Q/q intensity ratios that could be used for the reliable confirmation of compounds in


70

samples. The tolerances accepted were in accordance with the European Commission

Decision (2002/657/CE).28

Agreement in the retention times of the analyte in sample and of

the reference standard was also required to confirm a positive result (relative error ±0.5%).

Analytical parameters

Blank samples used for validation purposes contained appreciable concentrations of

several PAHs. Therefore, it was necessary to accurately calculate this concentration in order

to correct the quantitative results in recovery experiments.27

One labelled standard

(benzo[a]anthracene-D12) was added at the initial stage of the procedure as a quality

control (surrogate) in order to correct for possible losses during the overall procedure and

instrumental deviations for all the compounds except for naphthalene and acenaphthylene

which were processed without a surrogate.

Fortified samples of the four matrices studied in this work were analyzed applying

the developed methodology with satisfactory results (see Tables 3 and 4).

Fish fillet and fish feed

Pools of fish fillet and fish feed, blending 325 g of left-hand fillets and 50 g of fish

feed, respectively, were used for the statistical validation experiments. First, six replicates

of each matrix were analyzed by applying the analytical procedure proposed to determine

the content of selected analytes in these samples. These showed the presence of trace

levels of most of them, as could be expected because of the usual presence of PAHs in

marine samples,2,3,5

especially in fish feed due to the concentration of contaminants during

the manufacturing process with fish derivatives.35

Fluorene, phenanthrene, fluoranthene

and pyrene presented the highest concentrations in these blank samples. As can be seen in

Table 3, samples were fortified at 0.125, 1.25 and 2.5 µg/kg (n = 6) and submitted to the

developed procedure. In general, recoveries were satisfactory, with average values

between 80 and 110% and RSDs of less than 30%, except for naphthalene which presented

poor precision. LODs were found to be at sub-µg/kg level in both sample matrices (all 0.1


71

µg/kg). Such low LODs were achieved thanks to the efficient cleanup applied, together with

the selectivity and sensitivity provided by QqQ-MS/MS in SRM mode.

The average Q/q intensity ratios calculated from reference standards in solvent (see

Table 2) were compared with those experimentally obtained from spiked sample extracts to

test the robustness of these values and for the presence of potential matrix interferences

that might affect Q/q ratios and, consequently, the confirmation process. The average

deviations obtained were in all cases in accordance with the criteria indicated in the

validation section.

As an example, Fig. 2 shows representative chromatograms of fish fillet

(naphthalene, dibenzo[a,h]anthracene) and fish feed (acenaphthene and

benzo[g,h,i]perylene) fortified at 0.125 µg/kg (except for naphthalene, 1.25 µg/kg).

Q/q ratios for all analytes were between 1 and 5, except for pyrene, which showed

by far the highest Q/q ratio with a value of 28.17 (Table 2). This allows the determination of

pyrene at the LOQ level, but makes its confirmation difficult at low concentrations below

the LOQ. No other transitions with the selected precursor ion (m/z 202) or with other

precursor ions were found to lead to better sensitivity.

Fish and Linseed oil

Frequently, oils added in fish feed, such as fish oil and linseed oil, are a source of

contamination.36

These ingredients are the main part of the fish feed compositions,

necessary to reach a healthy composition of the diets (Table 1). For these reasons, data

obtained in the analysis of the blank sample used for validation reveal appreciable

concentrations of PAHs. As reported by several authors, fats and oils represent one of the

major sources of contamination in the diet because of their lipophilic nature.37

Table 4

shows the analytical parameters for the oil matrices studied, with validation data expressed

as µg/kg lipid weight. Fluorene, phenanthrene, fluoranthene and pyrene were, similarly to

fish fillet and fish feed, the most abundant PAHs found in the oils. Blanks were fortified at

1.25 and 2.5 µg/kg and submitted to the developed procedure with satisfactory recoveries

in all cases, and with a RSD lower than 30%. The limits of detection were at the low µg/kg


72

levels in fish oil and linseed oil. The LODs, LOQs and Q/q ratios were calculated in the same

way as for the fish fillet and fish feed, obtaining in all cases satisfactory results in

accordance with the criteria established.

Figure 2. Blank sample, spiked sample (0.125 µg/kg) and standard (1 ng/mL, equivalent to 0.125

µg/kg in sample) GC/MS/MS chromatograms from fish fillet and fish feed. SRM transitions

monitored: naphthalene (m/z 128 > 102), dibenzo[a,h]anthracene (m/z 278 > 276), acenaphthene

(m/z 152 > 126) and benzo[g,h,i]perylene (m/z 276 > 274). *Spiking level for naphthalene in fish fillet

was 1.25 µg/kg, and the reference standard was 10 ng/mL, equivalent to 1.25 µg/kg in sample.


73

Application to real samples

The optimized sample procedure followed by GC/(QqQ)MS/MS and GC/TOFMS was

applied (in triplicate) to the analysis of raw materials, fish feeds and fillet samples from

AQUAMAX long-term feeding trials with gilthead sea bream (Sparus aurata L.) as part of a

European project. Fish were exposed through the entire productive cycle to experimental

diets with graded levels of fish oil replacement, studying the health and welfare of the

farmed fish, and maximizing the health-promoting properties, safety, quality and

acceptability of the final product to the consumer.35

PAHs were determined in 19 fish fillet

samples, 8 fish feeds, 1 fish oil, 1 linseed oil, 1 rapeseed oil and 1 palm oil from fish exposed

through the productive cycle (14 months) to experimental diets with different percentages

of fish oil replacement with vegetable oils.

For the GC/QqQ-MS analysis, Table 6 shows the PAH concentrations detected in

aquaculture samples. All PAHs studied were found in fish feed in the concentration range of

0.2-12.7 µg/kg. The only exception was naphthalene in six fish feed samples, where it was

found at concentrations around 200 µg/kg fresh weight. All fish fillet samples analyzed

were positive for phenanthrene, fluoranthene and pyrene (range of 0.2-11.4 µg/kg).

Benzo[a]pyrene was detected only in one sample of fish fillet, but at a concentration

(3.9µg/kg) above the maximum established by the European Commission Regulation (EC)

No 1881/2006. Rapeseed and palm oils presented lower total loads of PAHs (12 µg/kg) than

fish and linseed oils, which had total loads of 65.7 and 47 µg/kg, respectively, principally

due to the major presence of fluorene, phenanthrene, fluoranthene and pyrene (see values

in Table 4).

Illustrative chromatograms for real samples analyzed are shown in Fig. 3, where the

quantification and confirmation transitions monitored for several PAHs can be seen in the

different matrices analyzed. As shown in the chromatograms, both quantification and

confirmation were feasible at sub-µg/kg levels with satisfactory peak shape.

Benzo[a]pyrene was not detected in fish fillet, whereas it was quantified and confirmed in

the diets.


74

Figure 3. GC/MS/MS chromatograms for selected PAHs in real aquaculture samples. Q quantification

transition, q confirmation transition. : Q/q ratio within tolerance limits. SRM transitions monitored:

acenaphthene (m/z 154 > 152, 153 > 126), pyrene (m/z 202 > 200, 202 > 152), benzo[a]pyrene (m/z

252 > 250, 250 > 248), fluoranthene (m/z 202 > 200, 202 > 150), benzo[a]anthracene, (m/z 228 >

226, 228 > 224).

GC/TOFMS was used for the additional confirmation of PAHs previously detected by

QqQ-MS. The GC/TOF detection and identification of target PAHs in the samples were

carried out by obtaining up to four narrow-window eXtracted Ion Chromatograms (nw-XICs)

(0.02 m/z units) at selected m/z values for every compound (Table 5). The TargetLynx

software automatically processed data and reported qualitative data. Analyte confirmation

was carried out by comparison of the Q/q intensity ratios obtained in samples with those

obtained from reference standards in solvent. In all cases, the presence of chromatographic

peaks at the expected retention time and the attainment of all Q/q ratios when comparing

with the reference standard allowed the confirmation of these findings in samples. In


75

addition, EI-TOFMS accurate mass spectra were obtained and the mass errors for

representative ions were calculated, giving high confidence in the confirmation process.

As an example, Fig. 4 shows GC/TOF chromatograms (nw-XICs) for pyrene and

naphthalene detected in fish fillet and fish feed. Reliable confirmation was feasible as all

the Q/q ratios were in agreement with the European Commission Decision (2002/657/CE).28

In addition, EI accurate mass spectra provided mass errors for the four ions monitored of

below 2.9 m m/z units.

Most of the positives found by triple quadrupole MS could be confirmed by TOFMS.

When the findings could not be confirmed by TOFMS, the reason was the absence of the

chromatographic peak in the nw-XIC. All these cases occurred when the analyte

concentration was very low (generally 1 µg/kg), surely as a consequence of the lower

sensitivity of the TOF mass spectrometer than of the triple quadrupole instrument working

in SRM mode.

However, in the full-analysis field, the capability of the TOF mass spectrometer to

allow interrogation of full-spectrum data after acquisition is interesting, as it allows us to

screen for unexpected compounds and metabolites at the time of injection without re-

analyzing the sample.38,39

The main advantage is the huge number of compounds that might

be investigated, with the obvious restrictions deriving from the requirements of GC and MS

analysis. High-occurrence metabolites, or new compounds that have been properly

identified by TOFMS, might then be included in analytical methods using QqQ-MS if the

commercial reference standards are available. This attractive approach is under study at

present in our laboratories for the samples analyzed in this work.


76

Table 6. PAH concentrations in aquaculture samples analyzed by the application of the developed method.

Fish Fillet(a)

Fish Feed(b)

Rapeseed Oil(c)

Palm Oil(d)

Compound Positives conc. range

(µg/kg) Positives

conc. range

(µg/kg) Positives

conc. range

(µg/kg) Positives

conc. range

(µg/kg)

Naphthalene 2 2.4-3.9 7 2.9-242 1 1.9 1 0.8

Acenaphthylene 1 0.2 7 0.2-1.6 1 0.2 1 0.2

Acenaphthene 1 0.2 7 0.2-1.7 1 0.6 1 0.5

Fluorene 18 0.2-0.8 7 1.7-6.4 1 0.9 1 0.4

Phenanthrene 19 0.5-3.7 8 0.4-12.7 1 1.2 1 1.3

Anthracene 2 0.6-1 8 0.4-2.1 1 1.6 - -

Fluoranthene 19 0.2-3.9 7 2-4 - - - -

Pyrene 19 0.2-11.4 8 0.8-7 - - - -

Benzo(a)anthracene 6 0.06-4.9 8 0.4-0.9 1 0.9 1 1.2

Chrysene 12 0.1-6.9 8 0.6-1.5 1 0.8 - -

Benzo(b)fluoranthene 1 5.9 8 0.6-2.7 1 0.4 1 1.3

Benzo(k)fluoranthene 1 5.9 8 0.3-2.3 1 0.4 1 1.3

Benzo(a)pyrene 1 3.9 2 0.45-0.5 1 0.7 1 1.4

Indeno (1,2,3-cd)pyrene - - 1 0.3 1 0.7 1 1.1

Dibenzo(a,h)anthracene - - 1 0.2 1 0.9 1 1.3

Benzo(g,h,i)perylene 1 3.4 1 0.4 1 0.8 1 1.2

Concentrations detected from fish oil and linseed oil are presented in Table 4. Total number of samples analyzed: (a) 19, (b) 8, (c) 1 and (d) 1.

Capítulo 4. Determinación de PAHs mediante GC-MS/MS y GC-TOF MS

77

Figure 4. GC/TOFMS extracted ion chromatograms at four m/z values (mass window 0.02 m/z units)

for pyrene detected in fish fillet and naphthalene detected in fish feed. Accurate mass spectra

showing the most relevant ions together with the experimental mass errors, in m m/z units

(bottom). : Q/q ratio within tolerance limits. St: reference standard; S: sample. Ions selected are

shown in Table 5.

CONCLUSIONS

A rapid, sensitive and selective analytical methodology for the determination of PAHs

in high lipid content aquaculture samples has been developed, reaching low quantification

levels by means of an efficient cleanup step combining saponification and SPE procedures


78

prior to the injection of the extracts into a gas chromatograph coupled to QqQ and TOF

mass spectrometers. Especial attention was paid to naphthalene, the earliest eluting

chromatographic peak, which was a major problem in part due to the high levels found in

the blank fish fillet and fish feed samples used for validation experiments. The combined

use of triple quadrupole and time-of-flight analyzers gives an extraordinary reliability to the

confirmation process of the compounds detected in samples.

ACKNOWLEDGEMENTS

The authors are thankful for the financial support of the European Union Project

Sustainable Aquafeeds to Maximise the Health Benefits of Farmed Fish for Consumers

(AQUAMAX; Contract number: 016249-2). The authors acknowledge the Serveis Centrals

d'Instrumentació Científica (SCIC) of University Jaume I for use of the GC/TOFMS system.

REFERENCES

1.Menzie CA, Potocki BB, Santodonato J. Environ. Sci. Technol. 1992; 26: 1278.

2.Baumard P, Budzinski H, Garrigues P, Sorbe JC, Burgeot T, Bellocq J. Mar. Pollut. Bulletin

1998; 36: 951.

3.Filipkowska A, Lubecki L, Kowalewska G. Anal. Chim. Acta 2005; 547: 243.

4.Hellou J, Leonard J, Collier TK, Ariese F. Mar. Pollut. Bull. 2006; 52: 433.

5.Moret S, Conte LS. J. Chromatogr. A 2000; 882: 245.

6.Perugini M, Visciano P, Giammarino A, Manera M, Di Nardo W, Amorena M.

Chemosphere 2007; 66: 1904.

7.USEPA. Toxicological profile for polycyclic aromatic hydrocarbons (PAHs). Available:

http://www.usepa.gov/S.

8.Martí-Cid R, Bocio A, Llobet JM, Domingo JL. Food Chem. Toxicol. 2007; 45: 1968.

9.Martí-Cid R, Llobet JM, Castell V, Domingo JL. Food Chem Toxicol. 2008; 46: 3163.

Capítulo 4. Determinación de PAHs mediante GC-MS/MS y GC-TOF MS

79

10.Commission Regulation (EC) No 1881/2006 of 19 December 2006 setting maximum

levels for certain contaminants in foodstuffs. Available:

eurlex.europa.eu/LexUriServ/site/en/oj/2006/l_364/l_36420061220en00050024.pdf.

11.Baumard P, Budzinski H, Garrigues P, Narbonne JF, Burgeot T, Michel X, Bellocq J. Mar.

Environ. Res. 1999; 47: 415.

12.Hellou J, Haya K, Steller S, Burridge L. Aquat. Conserv. 2005; 15: 349.

13.Malik A, Ojha P, Singh KP. B. Environ. Contam. Tox. 2008; 80: 134.

14.Hellou J, Haya K, Steller S, Burridge L. Aquat. Conserv. 2005; 15: 349.

15.Patel K, Fussell RJ, Hetmanski M, Goodall DM, Keely BJ. J. Chromatogr. A 2005; 1068: 289.

16.Poster DL, Schantz MM, Sander LC, Wise SA. Anal. Bioanal Chem. 2006; 386: 859.

17.Pena T, Pensado L, Casais C, Mejuto C, Phan-Tan-Luu R, Cela R. J. Chromatogr. A 2006;

1121:163.

18.Navarro P, Cortazar E, Bartolomé L, Deusto M, Raposo JC, Zuloaga O, Arana G, Etxebarria

N. J. Chromatogr. A 2006; 1128: 10.

19.Liguori L, Heggstad K, Hove HT, Julshamn K. Anal. Chim. Acta 2006; 573-574: 181.

20.Fernández-González V, Concha-Graña E, Muniategui-Lorenzo S, López-Mahía P, Prada-

Rodríguez D. J. Chromatogr. A 2007; 1176: 48.

21.Fernández-González V, Muniategui-Lorenzo S, López-Mahía P, Prada-Rodríguez D. J.

Chromatogr. A 2008; 1207: 136.

22.Campíns-Falcó P, Verdú-Andrés J, Sevillano-Cabeza A, Molins-Legua C, Herráez-

Hernández R. J. Chromatogr. A 2008; 1211: 13.

23.Loutfy N, Fuerhacker M, Tundo P, Raccanelli S, Ahmed MT. Chemosphere 2007; 66: 1962.

24.Pitarch E, Medina C, Portolés T, López FJ, Hernández F. Anal. Chim. Acta 2007; 583: 246.

25.Portolés T, Pitarch E, López FJ, Sancho JV, Hernández F. J. Mass Spectrom. 2007; 42:

1175.

26.CăIjka T, Hajlová J. J. Chromatogr. A 2004; 1058: 251.

27.Hernández F, Portolés T, Pitarch E, López FJ, Beltrán J, Vázquez C. Anal. Chem. 2005; 77:

7662.


80

28.Commission Decision 2002/657/EC implementing Council Directive 96/23/EC concerning


eur- lex.europa.eu/pri/en/oj/dat/2002/l_221/l_221200 20817en00080036.pdf.

29.Hernández F, Sancho JV, Pozo OJ. Anal. Bioanal. Chem. 2005; 382: 934.

30.Ferrer I , Thurman EM (eds). Liquid Chromatography Mass Spectrometry/Mass

Spectrometry, MS/MS and Time-of-Flight MS: Analysis of Emerging Contaminants,

American Chemical Society Symposium Series 850, Oxford University Press: New York,

2003.

31.Numata M, Yarita T, Aoyagi Y, Yamazaki M, Takatsu A. Chemosphere 2005; 58: 865.

32.Baumard P, Budzinski H, Garrigues P. Polycycl. Aromat. Comp. 1996; 9: 241.

33.Barranco A, Alonso-Salces RM, Bakkali A, Berrueta LA, Gallo B, Vicente F, Sarobe M. J.

Chromatogr. A 2003; 988: 33.

34.Santos FJ, Galceran MT. TrAC - Trends Anal. Chem. 2002; 21: 672.

35.Nácher-Mestre J, Serrano R, Benedito-Palos L, Navarro JC, Lopez FJ, Pérez-Sanchez J.

Chemosphere 2009; accepted.

36.Dennis MJ, Massey RC, Cripps G, Venn I, Howarth N, Lee G. Food Addit. Contam. 1991; 8:

517.

37.Van der Lee MK, van der Weg G, Traag WA, Mol HGJ. J. Chromatogr. A 2008; 1186: 325.

38.Hernández F, Portolés T, Pitarch E, López FJ. Anal. Chem. 2007; 79: 9494.

39.Hernández F, Portolés T, Pitarch E, López FJ. J. Mass Spectrom. 2009; 44: 1.


81

4.2.3. Discusión de los resultados.

La USEPA propuso una lista de 16 PAHs como contaminantes prioritarios debido a su

potente toxicidad sobre la población y su presencia en lugares de riesgo. En nuestro caso, la

elección de los PAHs a analizar se llevó a cabo considerando la lista propuesta por la EPA

por incluir aquellos PAHs utilizados como marcadores de bioacumulación en muestras de

pescado (EPA 1993)(EPA 2000). Asimismo, la lista de la EPA considera PAHs con 2, 3, 4, 5 y

6 anillos de benceno con características carcinogénicas y mutagénicas. Del mismo modo, se

incluye al naftaleno, que es considerado como un contaminante altamente ubicuo y

encontrado en concentraciones elevadas en sedimentos y tejidos de peces (Liang et al.,

2007). La presencia de fenantreno, antraceno, fluoranteno y pireno en esta lista también

fue un factor determinante ya que suelen presentar valores del orden de µg/Kg en

muestras de origen marino (Cheung et al., 2007; Berntssen et al., 2010).

La autoridad europea “European Food Safety Authority” (EFSA) publicó en 2002 una

lista con 15 PAHs prioritarios para su control y determinación en muestras alimentarias.

Sólo algunos de estos PAHs estaban incluidos previamente en la lista propuesta por la EPA y

se incluyen otros por su toxicidad (EFSA 2002). En 2005, las organizaciones “Food and

Agriculture Organization” (FAO) y la “World Health Organization” (WHO) añadieron a la lista

de la EFSA un nuevo PAH considerado como tóxico: “benzo[c]fluorine” (FAO/WHO, 2005)

con el que la lista quedaba ampliada a 16 PAHs pero fue nombrada como “15+1” (15+1 PAH

list) para diferenciarla de la lista de la EPA (16 PAH list). Aún así, en todos los casos sigue

tomándose como referencia la presencia de benzo(a)pireno, incluido en ambas listas como

marcador de la toxicidad por PAHs.

Los 16 PAHs incluidos en la lista de la EPA fueron seleccionados en nuestra

investigación como indicadores de este grupo de contaminantes en organismos marinos.

Estudios recientes, como el realizado por Marti-Cid (2007), sobre la ingesta de PAHs debido

al consumo de pescado y mariscos contaminados por niños (Martí-Cid et al., 2007), se han

basado también en el análisis de los 16 PAHs propuestos por la EPA.


82

• Optimización del método de extracción/purificación mediante saponificación

Para la extracción de los PAHs en muestras complejas, siguen utilizándose hoy en día

técnicas de extracción tradicionales como el reflujo y Soxhlet debido a su buen

funcionamiento. Actualmente también se utilizan técnicas automatizables. Como se ha

comentado en la introducción de esta Tesis, técnicas de extracción como la PLE, ASE, MAE y

la SPME (Schantz et al., 2006; Fidalgo-Used et al., 2007) muestran también buenos

resultados (van Leeuwen et al., 2008; Björklund et al., 2006). Yusà et al. (2005) propuso una

lista general de procedimientos analíticos para la determinación de PAHs en muestras de

pescado y marisco. En esta lista se incluyen diferentes procesos de extracción/purificación

mediante saponificación (Yusà et al., 2005). Tras la extracción, es frecuente la necesidad de

una etapa adicional de purificación del extracto, con el fin de obtener los mejores

parámetros analíticos en la determinación instrumental. En nuestro caso, el alto contenido

en grasa hizo necesario la optimización de un exhaustivo tratamiento de la muestra previo

al análisis por GC-MS/MS y GC-TOF MS.

En primer lugar se aplicó una etapa de saponificación para eliminar la gran cantidad

de grasa presente en las muestras. La saponificación resultó ser eficaz por eliminar un

elevado porcentaje de grasa y no degradar los PAHs. Como los PAHs son ácido-lábiles no se

pudo realizar el tratamiento con ácido por observarse degradación de los mismos. El

tratamiento con ácido es muy eficaz para eliminar grasa y es utilizado frecuentemente en

metodologías analíticas para la determinación de PBDEs y PCBs (Covaci et al., 2003; Serrano

et al., 2003). Las condiciones de saponificación se optimizaron con el objeto de eliminar la

mayor cantidad posible de ácidos grasos derivados de los lípidos de las matrices. Para ello,

se ensayaron diferentes temperaturas, disolventes y tiempos de reacción. Se realizaron

diferentes experimentos que incluían la mezcla, agitación y posterior extracción líquido-

líquido: saponificación inmediata, saponificación durante una hora, dos y tres horas en

condiciones de temperatura ambiente, 60°C y 80°C. Para todas ellas se utilizaron los

alcalinizantes KOH y NaOH en diferentes disolventes: agua, EtOH, MeOH, preparando

disoluciones a diferentes concentraciones y volúmenes: 1M, 2M y saturada. Además se

añadió NaCl, agua y hexano individualmente para mejorar la separación de las fases.


83

Los mejores resultados se obtuvieron realizando la saponificación durante tres horas

a una temperatura de 80°C como se puede observar en la Figura 1 (Artículo científico 1).

• Optimización de la purificación mediante SPE

En nuestro caso la saponificación no fue suficiente para obtener extractos lo

suficientemente limpios como para conseguir resultados óptimos en la determinación

mediante GC-MS/MS, por lo que se aplicó una etapa adicional de purificación basada en

SPE.

Otros autores también han propuesto metodologías para la determinación de PAHs

acoplando diferentes técnicas de purificación dado que, en muestras complejas, una única

etapa de purificación suele resultar insuficiente para determinar cuantitativamente los

analitos a niveles de concentración suficientemente bajos (Fromberg et al., 2007).

A pesar de los avances en la separación y la detección de los sistemas

cromatográficos, la etapa de tratamiento de muestra resulta trascendental para obtener

unos datos fiables. Hoy en día todavía existe una necesidad de búsqueda de metodologías

de tratamiento de muestras complejas, que sean eficaces, económicas y rápidas para la

determinación de compuestos orgánicos arraigados a matrices con alto contenido graso

(Gilbert-López et al., 2009).

Entre las técnicas ampliamente utilizadas hoy en día, la SPE es utilizada

frecuentemente por presentar grandes posibilidades en la purificación y concentración de

las muestras con el fin de obtener una gran sensibilidad en la medida analítica. Varios

autores han utilizado la SPE en metodologías analíticas para la determinación de

contaminantes orgánicos, incluyendo PAHs, en muestras medioambientales a niveles de

ng/L (Matamoros et al., 2010). Otra técnica ampliamente utilizada es la GPC ya que es una

potente herramienta de separación y concentración de los analitos en muestras complejas.

Este tipo de técnica se utiliza frecuentemente para separar compuestos de tamaño

molecular elevado de la matriz, como por ejemplo los lípidos, ofreciendo resultados

satisfactorios (Fromberg et al., 2007; Jira et al., 2008).


84

En nuestro caso, la aplicación de SPE a la purificación de los extractos de las

diferentes matrices consideradas mostró una elevada eficacia en la eliminación de los

lípidos y permitió la concentración de los PAHs, alcanzando LOQs a niveles de µg/Kg.

Para ello, fue indispensable optimizar el proceso de SPE, ensayando diferentes fases

estacionarias, disolventes y procedimientos de elución. Hay que considerar que,

dependiendo del disolvente empleado, el patrón de elución de los PAHs variaba

considerablemente. Así se probaron diferentes disolventes, entre ellos el hexano,

diclorometano, acetato de etilo y mezclas entre ambos. Al final se decidió utilizar una

mezcla de diclorometano/hexano (20:80) como mejor disolvente para la elución de los

PAHs debido a la elución correcta de los PAHs manteniendo la fracción lipídica retenida en

la fase estacionaria de Florisil.

• Optimización de GC-MS/MS

Con el objetivo de conseguir LOQs lo más bajos posible, se empleó la GC acoplada a

un analizador de masas QqQ. Este analizador permite trabajar en MS/MS obteniendo un

alto grado de fiabilidad en la identificación de los analitos. En modo de adquisición “SRM”,

los analizadores QqQ resultan excelentes para la cuantificación y confirmación de la

identidad de los compuestos a niveles traza, especialmente en matrices complejas. Para la

determinación de PAHs se seleccionaron dos transiciones por compuesto atendiendo a su

mayor sensibilidad. A raíz de la pobre fragmentación de los PAHs, el pico molecular fue

escogido como precursor para la transición de cuantificación en la mayoría de los casos

(Tabla 2, Figura 3. Artículo Científico 1).

• Optimización de GC-TOF MS

GC-TOF MS se utilizó para realizar una confirmación de la identidad adicional de los

PAHs detectados con el GC-MS/MS. Se seleccionaron cuatro iones para la identificación de

cada uno de los compuestos analizados. La relación m/z más intensa fue seleccionada para

la cuantificación (Q) mientras que las tres restantes fueron adquiridas para la confirmación


85

(q) y estudio de las relaciones Q/q (Tabla 5, Figura 4. Artículo de Investigación 1). El

estudio de los valores Q/q se realizó atendiendo al reglamento emitido por la Unión

Europea sobre metodología analítica e interpretación de resultados (Commission Decision

2002/657/EC).

• Validación de la metodología analítica

La validación de la metodología analítica se desarrolló para cuatro matrices (filete de

dorada cultivada, pienso, aceite de pescado y aceite de lino). Se llevó a cabo mediante

experimentos de recuperación en muestras blanco seleccionadas fortificadas a tres niveles

de concentración en filetes de pescado y pienso (0.125, 1.25 and 2.5µg/Kg, n=6) y dos

niveles en aceite de pescado y aceite de lino (1.25 and 2.5µg/Kg, n=3). En general se

obtuvieron recuperaciones satisfactorias (80-110%), así como valores óptimos de precisión

(RSD<30%). Naftaleno, fenantreno, fluoranteno y pireno presentaron concentraciones

apreciables en el blanco por lo que fueron validados a 1.25 µg/kg.

No se encontraron materiales de referencia certificados de las matrices consideradas

a los niveles de concentración más bajos estudiados en el presente trabajo (0.125µg/kg en

filete y pienso; 1.25 µg/kg en aceites) por lo que la validación estadística mediante matrices

“blanco” fortificadas a diferentes concentraciones fue seleccionada como la mejor opción

considerando todos los parámetros propuestos por la guía SANCO vigente (Document N°

SANCO/2007/3131).

4.2.4. Referencias.

Aquamax Issue N° 01. 2008. Aquamax Project: Sustainable Aquafeeds to Maximise the

Health Benefits of Farmed Fish for Consumers. www.aquamaxip.eu.


86

Bartle, K.D. 1991. En “Food contaminants: Sources and Surveillance” (Creaser C, Purchase R.

eds). The Royal Society of chemistry, Cambridge. 41-60.

Berntssen, M.H.G.; Julshamn, K.; Lundebye, A. 2010. Chemical contaminants in aquafeeds

and Atlantic salmon (Salmo salar) following the use of traditional- versus alternative

feed ingredients. Chemosphere 78, 637-646.

Cheung, K.C.; Leung, H.M.; Kong, K.Y.; Wong, M.H. 2007. Residual levels of DDTs and PAHs

in freshwater and marine fish from Hong Kong markets and their health risk

assessment. Chemosphere 66, 460-468.

Chou, P.; Matsui, S. 2006. Contamination of polycyclic aromatic hydrocarbons and their

implication for environmental hazard. Water Sci. Technol. 6, 165-174.

Commission Regulation 1881/2006/EC of 19December 2006 setting maximum levels for

certain contaminants in foodstuffs. Available: http://eur-

lex.europa.eu/LexUriServ/LexUriServ.do?uri=OJ:L:2006:364:0005:0024:EN:PDF



http://eur-


Covaci, A.; Voorspoels, S.; de Boer, J. 2003. Determination of brominated flame retardants,

with emphasis on polybrominated diphenyl ethers (PBDEs) in environmental and

human samples. Environ. Int. 29, 735-756.

Document N° SANCO/2007/3131. 2007. Method validation and quality control procedures

for pesticide residues analysis in food and feed.

EFSA. 2002. Opinion of the Scientific Committee on Food on the Risks to Human Health of

Polycyclic Aromatic Hydrocarbons in Food, 2002, SCF/CS/CNTM/PAF/29/Final.


87

EPA. 1993. Provisional guidance for quantitative risk assessment of polycyclic aromatic

hydrocarbons. EPA 600/R-93/089. US Environmental Protection Agency, Washington

DC.

EPA. 2000. Guidance for assessing chemical contaminant data for use in fish advisories. EPA.

823-B-00-008. US Environmental Protection Agency, Washington DC.

FAO/WHO. 2005. Summary and Conclusions of the Joint FAO/WHO Expert Committee on

Food Additives, Sixty-Fourth meeting, Rome, 2005, JECFA/64/SC.

Fromberg, A.; Højgård, A.; Duedahl-Olesen, L. 2007. Analysis of polycyclic aromatic

hydrocarbons in vegetable oils combining gel permeation chromatography with solid-

phase extraction clean-up. Food Addit. Contam. 24, 758-767.

Gilbert-López, B.; García-Reyes, J.F.; Molina-Díaz, A. 2009. Sample treatment and

determination of pesticide residues in fatty vegetable matrices: A review. Talanta. 79,

109-128.

Jira, W.; Ziegenhals, K.; Speer, K. 2008. Gas chromatography-mass spectrometry (GC-MS)

method for the determination of 16 European priority polycyclic aromatic

hydrocarbons in smoked meat products and edible oils. Food Addit. Contam. A 25,

704-713.

Koss, G. & I. Tesseraux. (1999) En "Toxicology" (Marquardt H. et al., eds). Academic Press,

San Diego. 603-644

Liang, Y.; Tse, M.F.; Young, L.; Wong, M.H. 2007. Distribution patterns of polycyclic aromatic

hydrocarbons (PAHs) in the sediments and fish at Mai Po Marshes Nature Reserve,

Hong Kong. Water Res. 41, 1303-1311.

Lin, D.; Tu, Y.; Zhu, L. 2005. Concentrations and health risk of polycyclic aromatic

hydrocarbons in tea. Food Chem. Toxicol. 43, 41-48.


88

Mar^-Cid, R.; Bocio, A.; Llobet, J.M.; Domingo, J.L. 2007. Intake of chemical contaminants

through fish and seafood consumption by children of Catalonia, Spain: Health risks.

Food Chem. Toxicol. 45, 1968-1974.

Matamoros, V.; Jover, E.; Bayona, J.M. 2010. Part-per-Trillion determination of

pharmaceuticals, pesticides, and related organic contaminants in river water by solid-

phase extraction followed by comprehensive two-dimensional gas chromatography

time-of-flight mass spectrometry. Anal. Chem. 82, 699-706.




J. Sep. Sci. 26, 75-86.

Srogi, K. 2007. Monitoring of environmental exposure to polycyclic aromatic hydrocarbons:

A review. Environ. Chem. Lett. 5, 169-195.

Talsness, C.E. 2008. Overview of toxicological aspects of polybrominated diphenyl ethers: A

flame-retardant additive in several consumer products. Environ. Res. 108, 158-167.

United States Environmental Protection Agency, Polynuclear Hydrocarbons, vol.

10, US-EPA, Washington, DC, 1984.

Van der Oost, R.; Beyer, J.; Vermeulen, N.P.E. 2003. Fish bioaccumulation and biomarkers in

environmental risk assessment: A review. Environ. Toxicol. Pharmacol. 13, 57-149.

Vonderheide, A.P.; Mueller, K.E.; Meija, J.; Welsh, G.L. 2008. Polybrominated diphenyl

ethers: Causes for concern and knowledge gaps regarding environmental distribution,

fate and toxicity. Sci. Total Environ. 400, 425-436.

Yusà, V.; Pardo, O.; Martí, P.; Pastor, A. 2005. Application of accelarated solvent extraction

followed by gel performance chromatography and high-performance liquid

chromatography for the determination of polycyclic aromatic hydrocarbons in mussel

tissue. Food Addit. Contam. 22, 482-489.

Capítulo 4. Determinación de PBDEs mediante GC-MS

89

4.3. Metodología analítica para la determinación de difenil éter polibromados en

muestras complejas.


90


91


Existe una creciente preocupación como consecuencia de la presencia en el medio

ambiente de PBDEs debido a su alto poder de bioacumulación y a su toxicidad

(Vonderheide et al., 2008). La razón de la presencia de estos compuestos en diferentes

ambientes es debida a su uso como retardantes de llama en materiales que forman parte

de una gran variedad de artículos de consumo. Ejemplos de estos artículos serían las

espumas de poliuretano flexibles utilizadas en el acolchado de muebles, las carcasas de

aparatos eléctricos y electrónicos, principalmente en televisores y ordenadores. Los PBDEs

son liberados en el medio ambiente a través del polvo, por ordenadores personales,

televisores, alfombras, tapizados y a través de las aguas residuales (procedentes de

vertederos). En el medioambiente, los PBDEs entran a la cadena alimentaria y se pueden

encontrar en la grasa del ganado y los peces. Las rutas de exposición más comunes en seres

humanos son el consumo de comida contaminada (carnes, pescado, productos lácteos) o

respirar aire contaminado en el interior de casas o edificios (Frederiksen et al., 2009).

Además, los niños recién nacidos pueden estar expuestos a los PBDEs a través de la leche

materna contaminada (Talsness et al., 2008).

El término PBDEs engloba a todos los compuestos que presentan la fórmula

molecular recogida en la Figura 4.2. Existen 209 congéneres posibles según el número y la

posición de los átomos de bromo en la molécula de difenil éter. Los congéneres con el

mismo número de átomos de bromo en diferentes posiciones reciben el nombre de

homólogos, formando todos ellos la familia de homólogos correspondiente. Así, el término

PentaBDE hace referencia a todos los homólogos de difenil éter bromado con 5 átomos de

bromo.

Figura 4.2. Estructura molecular de los PBDEs.


92

Las propiedades físico-químicas más relevantes de estos compuestos son similares a

las de los PCBs, presentando baja presión de vapor y solubilidad, así como un alto valor de

la constante de reparto octanol/agua (expresada como log Kow), que determinan su

comportamiento ambiental. A medida que aumenta el grado de bromación, los valores de

presión de vapor y solubilidad disminuyen, aumentando el valor de log Kow. Este parámetro

nos advierte de su capacidad para acumularse en los organismos vivos y su potencial de

absorción sobre la materia orgánica de los suelos. Los contaminantes con valores de Kow

elevados (superiores a 4) tienen un potencial de bioacumulación muy alto, y por tanto, se

trata de compuestos que si se utilizan en cantidades elevadas pueden llegar a transferirse a

lo largo de las cadenas tróficas.

Como consecuencia de la necesidad de controlar la presencia de PBDEs en diferentes

matrices ambientales y, concretamente en los productos procedentes de la acuicultura

marina, el presente trabajo pretende desarrollar una metodología analítica para el análisis

de muestras reales de manera fiable y precisa tanto para la identificación como la

cuantificación de los PBDEs seleccionados. En el desarrollo de la metodología analítica se

prestó especial atención a la etapa de purificación de los interferentes presentes en las

muestras, debido a la problemática que presenta el análisis de extractos con un elevado

contenido lipídico.

Las matrices incluidas en el estudio son filetes de dorada cultivada, piensos con

diferente composición, aceite de pescado y aceites vegetales. Las muestras proceden de un

estudio experimental en el marco del proyecto AQUAMAX (Sustainable Aquafeeds to

Maximise the Health Benefits of Farmed Fish for Consumers. www.aquamaxip.eu).


93


Gas chromatography-mass spectrometric determination of polybrominated diphenyl ethers

in complex fatty matrices from aquaculture activities.

J. Nácher-Mestre, R. Serrano, F. Hernández, L. Benedito-Palos, J. Pérez-Sánchez.

Anal. Chim. Acta 664 (2010) 190–198

Anal. Chim. Acta 664 (2010) 190–198

94


95

Abstract

Gas chromatography coupled to mass spectrometry in negative chemical ionization mode

(GC-(NCI)MS) has been applied to the quantification and reliable identification of

polybrominated diphenyl ethers (PBDEs) in animal and vegetable samples from aquaculture

activities. Matrices analyzed included fish fillet, fish feed, fish oil and linseed oil, their fat

content ranged from 5% to 100%. Solid-phase extraction (SPE) (using Florisil and silica

cartridges) and normal-phase high performance liquid chromatography were tested for an

efficient clean-up in order to obtain sample extracts free of interfering compounds.

Combining sulphuric acid digestion and SPE with Florisil led to the highest efficiency in the

elimination of interferences from the extracts. The sample procedure developed, together

with the application of GC-(NCI)MS for measurement, led to the satisfactory determination

of PBDEs at μg kg −1 levels in complex aquaculture matrices with high lipid content. The use

of a short and thin film-thickness fused-silica capillary column allowed to determine the

problematic BDE 209 with satisfactory results. Three m/z ions were acquired for each

analyte, which ensured a reliable identification of compounds detected in samples.

Keywords: Polybrominated diphenyl ethers; Fatty samples; Gas chromatography; Mass

spectrometry; Clean-up; Aquaculture matrices

Anal. Chim. Acta 664 (2010) 190–198

96

1. Introduction

Polybrominated diphenyl ethers (PBDEs) are anthropogenic chemicals that are added

to a wide variety of consumer/commercial products in order to improve their fire

resistance. For this reason, PBDEs are commonly named as brominated flame retardants

(BFRs) due to their capability in reducing fire in a widespread list of products. BFR’s

frequent employment is to reduce successfully deaths related to the fire, injuries and

property damage [1]. This fact, together with their hydrophobic properties, contributes to

the presence and bioaccumulation of BFRs in environmental samples [2]. Analytical studies

on BFRs have been focused on environmental samples [3], as air, sewage sludge,

sediments, soils and water [4], and on biota for fish derivatives [5], and they are normally

found in biological tissues at µgkg−1 levels [6,7].

Marine aquaculture has experienced strong development in the last decades as a

consequence of increased fish consumption by the world population and decreasing wild

stocks. Fish used as raw material for manufacture of fish feed ingredients (fish oils and

meals) are a potential source of PBDEs in fish feed, which can be bioaccumulated by aquatic

species [5,8]. PBDEs’ determination in samples and resources from aquaculture is

important in order to establish the contamination origin, for a dietary exposure assessment

and for protection of public health [5,9,10].

The determination of PBDEs in marine matrices is difficult due to the low analyte

levels normally present and the presence of interfering compounds (e.g. fats), which are co-

extracted with the analytes. This makes an efficiency clean-up necessary before

instrumental measurement. Extraction of PBDEs in these samples can be carried out using

Soxhlet, sonication, matrix solid-phase dispersion, solid-phase microextraction or

pressurized liquid extraction [11-13]. The purification of fatty marine extracts is crucial to

get satisfactory analytical data and to reach the sensitivity required. An efficient clean-up

can be performed using gel permeation chromatography, solid-phase extraction (SPE) or

normal-phase high performance liquid chromatography (NPLC), as well as by acid treatment

or the use of acidified silica for lipid removal [13,14].

As regards analytical measurement, gas chromatography coupled to mass

spectrometry (GC-MS) is normally the preferred technique for PBDEs, although LC-MS


97

under different ionization modes has been also tested [1,15]. GC with electron capture

detector has been applied, but it has important limitations in selectivity because the

detection and identification is based only on retention time and matrix interferences can

appear [16]. Some authors have reported PBDEs measurements in environmental analysis

by GC-MS under selected ion monitoring mode (SIM) [17-19]. In the light of data reported,

one of the most sensitive approaches seems to be high resolution GC with negative

chemical ionization (NCI) coupled to a single quadrupole MS in SIM mode, while electron

ionization (EI) may improve the selectivity thanks to the most abundant fragmentation. NCI-

MS may be less specific than EI-MS due to the Br− ions formation, which can be confused

from interfering compounds in complex samples [16]. However, GC-(NCI)MS increases

sensitivity and accuracy in the determination of BDE 209, the most difficult BDE from an

analytical point of view, due to its particular behavior on degradation and sensitivity [20].

Analytical methodologies reported for the determination of PBDEs in marine fatty

samples are normally focused on tri, tetra, penta and hexa PBDEs congeners. Thus, BDE 47

(tetra-BDE), 99 and 100 (penta-BDEs), have been usually detected in this kind of samples

[21-23]. High-brominated compounds with seven or more bromine atoms have proven to

be more difficult to be determined correctly. Thus, high PBDEs, such as BDE 209, are not

often included in PBDE studies [24,25], possibly due to their tendency to degrade in

injectors and during gas chromatographic separation process [23]. Nevertheless, most of

the commercial mixtures of PBDEs reference standards also contain high-brominated

congeners, as they are relevant as a consequence of their higher lipophilic character and

consequently their greater tendency to be bioaccumulated in fatty animal tissues.

As regards the determination of PBDEs in fish feed and their ingredients, Fajar et al.

[26] proposed amethodology for the determination of seven PBDEs in fish feed and

molluscs, at µgkg−1 levels. However, fish and vegetal oils used in the fish feed manufacture

and farmed fish are normally scarcely investigated in the bibliography.

In this work, we have developed a sensitive method for the reliable quantification

and identification of 12 PBDEs (including tri, tetra, penta, hexa, hepta PBDEs congeners and

the deca congener BDE 209), in aquaculture samples with fat contents ranging from 5% to

100%. The method is based on the use of GC-MS working in NCI mode and selected ion

Anal. Chim. Acta 664 (2010) 190–198

98

recording (SIR). Working in SIR means that only the first quadrupole in the triple

quadrupole analyzer is acquiring. SIR is the nomenclature used by the manufacturer

(Waters) and it is equivalent to SIM, employed by other manufacturers. In order to obtain

extracts with very low content of fats and other interference compounds, the efficiency of

SPE and of NPLC as clean-up techniques has been studied after applying a first clean-up

step by sulphuric acid digestion. The developed methodology has been applied to real-

world samples of fish feed, vegetable and fish oils, and fish fillets from feeding trials with

gilthead sea bream (Sparus aurata L.) as a part of the European Union project “Sustainable

Aquafeeds to Maximise the Health Benefits of Farmed Fish for consumers” (AQUAMAX),

Contract number: 016249-2. The acquisition of three selective m/z ions for each analyte has

allowed quantification and reliable identification of PBDEs, at low µgkg−1 level, in the

different fatty samples analyzed.

2. Experimental

2.1. Materials and reagents

Standards of PBDEs (50 mg L−1 in isooctane) were obtained from Chiron As

(Trondheim, Norway) with purity higher than 99.5%: tri-BDE: BDE 28 (2,4,4'-

tribromodiphenyl ether), tetra-BDEs: BDE 71 (2,3',4',6-tetrabromodiphenyl ether), BDE 47

(2,2',4,4'-tetrabromodiphenyl ether), BDE 66 (2,3',4,4'-tetrabromodiphenyl ether), penta-

BDEs: BDE 100 (2,2',4,4',6-pentabromodiphenyl ether), BDE 99 (2,2',4,4',5-

pentabromodiphenyl ether), BDE 85 (2,2',3,4,4'-pentabromodiphenyl ether), hexa-BDEs:

BDE 154 (2,2',4,4',5,6'-hexabromodiphenyl ether), BDE 153 (2,2',4,4',5,5'-

hexabromodiphenyl ether), BDE 138 (2,2',3,4,4',5'-hexabromodiphenyl ether), hepta-BDE:

BDE 183 (2,2',3,4,4',5',6-heptabromodiphenyl ether) and the deca-BDE: BDE 209

(decabromodiphenyl ether). A standard mixture, which contained the 12 PBDEs (5 mg L−1)

was prepared by dilution of individual reference standard solutions with n-hexane, and it

was stored in a freezer at −20°C.

Working solutions of 12 PBDEs were prepared by diluting the standard mixture

solution with n-hexane, and they were used for sample fortification and for preparation of

calibration curves.


99

4,4-DDE D8 from Promochem (Wesel, Germany) was used as surrogate internal

standard (IS). Working solutions (100 µg L−1 and 250 µg L−1) were prepared by dilution of the

stock solution (100 mg L−1) with n-hexane and were stored at 4°C.

Ethyl acetate and n-hexane (ultratrace quality) were purchased from Scharlab

(Barcelona, Spain). Sulphuric acid 95-98% was purchased from Scharlab. Anhydrous sodium

sulphate of pesticide residue quality (Scharlab) was dried for 18 h at 300°C before use.

Strata silica cartridges (1 g) from Phenomenex (Torrance, CA, USA) and Supelclean LC-

Florisil SPE tube (1 g) from Sigma-Aldrich (Madrid, Spain) were used in SPE experiments.

2.2. Sample material

Gilthead sea bream (Sparus aurata L.) specimens of Atlantic origin (Ferme Marine de

Douhet, Ile d’Oléron, France) were cultured at the Instituto de Acuicultura de Torre la Sal,

Spain (IATS, CSIC) and collected when fish accomplished commercial size (≈500 g). The left-

side fillets (denuded from skin and bone) were excised and stored at −20°C until analysis.

Fish feed supplied to sea bream during feeding trials was stored at −20°C until analysis. Fish

oil and linseed oil used in fish feed manufacture, were also stored at −20°C until analysis.

2.3. NPLC clean-up procedure

NPLC clean-up system previously applied by Serrano et al. [14] was tested to remove

the interferences in the sample extracts. The system consisted of an LC Pump Master 305

from Gilson (Middleton, WI, USA), two VICI Valco six-way high pressure valves from Europe

instruments (Schenkon, Switzerland), a Rheodyne sampler injector valve from Rheodyne

(Colati, CA, USA) with 1mL loop, a Novapack silica column (150 mm×3.9 mm i.d., 4µm

particle size) from Waters (Mildford, MA, USA), and a Gilson FC 203B fraction collector. The

mobile phase was hexane at a flow rate of 1 mL min−1. The volume injected (sample

extracts and/or standard solutions) was 1 mL. The sample extract injected into the NPLC

system was previously treated with sulphuric acid to remove most of the fat content.

Firstly, 1-mL LC fractions after injecting PBDE standards were collected and analyzed

by GC-(NCI)MS in order to know their elution patterns in the LC clean-up. Results showed

that all PBDEs eluted in the fractions between 2 min and 8 min. Therefore, the whole

Anal. Chim. Acta 664 (2010) 190–198

100

fraction eluting from 2 min to 8 min was collected when injecting sample extracts and it

was adjusted to 0.25 mL by evaporation under N2 stream at 40°C. The overall procedure

was controlled from the LC pump, programming the times of valve activation automatically.

2.4. SPE clean-up procedure

SPE was also tested as clean-up procedure for fatty samples using Florisil and silica

cartridges. 1-mL of PBDEs mix standard solution was loaded into the cartridge, previously

conditioned with 6mL of n-hexane. Elution was carried with n-hexane and individual

fractions of 1mL were collected and analyzed by GC-(NCI)MS. Results showed that all PBDEs

eluted in the first 8mL using Florisil cartridges. In the case of silica SPE cartridges, 10mL n-

hexane was necessary for elution of all 12 PBDEs.

Once optimized the SPE-Florisil and SPE-silica clean-up procedures, 1-mL of the

cleaned-up hexanic extract sample (after acid treatment) was introduced in the SPE

cartridge (Florisil and silica, respectively) and fractions were collected. The whole eluate,

containing all analytes, was evaporated under a gentle nitrogen stream at 40°C, and the

final residue was dissolved in 0.25mL n-hexane in both cases and injected in the GC system.

2.5. GC instrumentation

A GC system (Agilent 6890N) from Agilent Technologies (Palo Alto, CA, USA)

equipped with an autosampler (Agilent 7683) was coupled to a triple quadrupole (QqQ)

mass spectrometer, Quattro Micro GC from Micromass (Boston, MA, USA), operating in NCI

mode. The GC separation was performed using a fused-silica DB-1HT capillary column with

a length of 15 m, an internal diameter of 0.25 mm and a film thickness of 0.1 μm from J&W

Scientific (Folsom, CA, USA). The injector temperature was set to 260°C. Splitless injections

of 1 μL were carried out. Helium (99.999%) from Praxair (Madrid, Spain) was used as carrier

gas at a flow rate of 1 mL min−1. The source temperature was set to 200°C and a solvent

delay of 3 min was selected. QqQ system operated in SIR mode. Methane (99.9995%) from

Praxair (Madrid, Spain) was used as reagent gas with an optimal flow of 60%. Experimental

conditions selected for GC-(NCI)MS measurement are shown in Table 1. The oven

temperature program was as follows: 120 °C (1 min); 10 °C min−1 to 220 °C; 20 °C min−1 to


101

300 °C (3 min), 40 °C min−1 to 340 °C (3 min). Helium was used as carrier gas at 1 mL min−1.

A dwell time per channel between 0.05 s and 0.1 s was chosen. The application manager

TargetLynx was used to process the quantitative data obtained from calibration standards

and samples.

2.6. Recommended procedure

Before analysis, samples were thawed at room temperature and carefully ground

using a mill Super JS from Moulinex (Bagnolet Cedex, France). Approximately 16 g of fish

fillet or fish feed was homogenized with the amount of anhydrous sodium sulphate

necessary to remove water (around 20 g Na2SO4), and the blend was spiked with 800 μL of

surrogate IS solution (250 μg L−1 4,4-DDE-D8). Extraction was performed by refluxing in n-

hexane for 4 h at 80 °C. After filtration (0.45 μm), the extract was pre-concentrated using a

Kuderna-Danish apparatus until ca. 8 mL (2 g sample per mL n-hexane). A sulphuric acid

digestion [14] was firstly applied to 1 mL of the hexanic extract, in order to remove most of

the fats.

In the case of oils, 0.5 g of fish oil or linseed oil was mixed with 1.5 mL of n-hexane

and was directly digested. Digestion was carried by adding 2 mL of conc. H2SO4 and shaking

1 min in a Vortex. The hexanic layer was collected, and the remaining sulphuric phase was

washed with 1 mL hexane twice. Finally, the around 3 mL cleaned-up hexanic phase was

adjusted to 1 mL at 40 °C under a gentle N2 stream.

Anal. Chim. Acta 664 (2010) 190–198

102

Table 1. GC-(NCI)MS measurement conditions selected for PBDEs determination.

PBDE number tR (min) Quantification Ion

(Q) (m/z)

Confirmation Ion (q)

(m/z) q Q/q1 Q/q2

Dwell

time (s)

p,p'-DDE-D8a 7.4 289 [C14D8

35Cl3]

-

0.1

28 7.9 79 [79

Br]-

81 [81

Br]- q1

1.01(5) >20 0.05 161 [H

79Br

81Br]

- q2

71 9.6 79 [

79Br]

-

81 [81

Br]- q1

1.01(1) >20 0.05 161 [H

79Br

81Br]

- q2

47 9.8 79 [

79Br]

-

81 [81

Br]- q1

1.01(1) >20 0.05 161 [H

79Br

81Br]

- q2

66 10.1 79 [

79Br]

-

81 [81

Br]- q1

1.05(4) >20 0.05 161 [H

79Br

81Br]

- q2

100 11.2 79 [

79Br]

-

81 [81

Br]- q1

1.02(1) >20 0.05 161 [H

79Br

81Br]

- q2

99 11.6 79 [

79Br]

-

81 [81

Br]- q1

0.99(7) >20 0.05 161 [H

79Br

81Br]

- q2

85 12.1 79 [

79Br]

-

81 [81

Br]- q1

0.98(10) >20 0.05 161 [H

79Br

81Br]

- q2

154 12.5 79 [

79Br]

-

81 [81

Br]- q1

0.99(8) >20 0.05 161 [H

79Br

81Br]

- q2

153 12.9 79 [

79Br]

-

81 [81

Br]- q1

0.96(13) >20 0.05 161 [H

79Br

81Br]

- q2

138 13.3 79 [

79Br]

-

81 [81

Br]- q1

0.96(9) >20 0.05 161 [H

79Br

81Br]

- q2

183 13.8 79 [

79Br]

-

81 [81

Br]- q1

1(7) >20 0.05 161 [H

79Br

81Br]

- q2

209 16.5 79 [

79Br]

-

81 [81

Br]- q1

0.98(6) 12.85(17) 0.05 487[C6O

79Br3

81Br2]

- q2

a Internal Standard used as surrogate.

b Average value calculated from standard solutions at seven concentration levels (n=2 each).

RSD (%) shown in brackets.


103

The 1-mL extract resulting from sulphuric acid digestion (neither pre-concentration

nor dilution of the sample extracts takes place) was passed through a 1 g Florisil SPE

cartridge, previously conditioned with 6 mL of n-hexane, and it was eluted with 8 mL n-

hexane. The eluate was evaporated under a gentle nitrogen stream at 40 °C and the final

residue was reconstituted in 0.25 mL of n-hexane. Along the whole analytical procedure 8-

fold pre-concentration (fish fillet and fish feed) or 2-fold pre-concentration (oil samples)

takes place.

The final cleaned-up extracts obtained were injected into the Quattro Micro GC

system working in NCI(MS) mode under the optimized conditions shown in Table 1.

Quantification of samples was carried out using calibration with standards in solvent, using

4,4-DDE D8 as surrogate IS for all congeners.

2.7. Validation

Validation of the method developed was performed by evaluating the following

parameters:

- Linearity. The calibration curves were obtained by injecting reference standard

solutions in n-hexane in duplicate. Linearity of relative response (analyte versus IS) was

tested by injecting standard solutions in duplicate. All PBDEs were studied in the

concentration range 0.5-50 μg L−1, while BDE 209 was tested between 2.5 μg L−1 and 250 μg

L−1 due to its lower sensitivity. Linearity was assumed when the regression coefficient was

greater than 0.99 with residuals randomly distributed and being lower than 30%.

- Accuracy. It was evaluated by means of recovery experiments, analyzing “blank”

reference samples (n = 3) of each matrix investigated, which were spiked at two

concentration levels: 0.25 μg kg−1 and 2.5 μg kg−1 (1.25 μg kg−1 and 12.5 μg kg−1 for BDE 209)

in fish fillet and fish feed, and 1 μg kg−1 and 10 μg kg−1 (5 μg kg−1 and 50 μg kg−1 for BDE 209)

in fish and linseed oils. Previously, several samples were analyzed in triplicate to determine

the concentration of the analytes in the different sample matrices. Data showed that BDE

28 and BDE 47 were most frequently detected in fish fillet and fish feed, while in oils PBDEs

were always below the limits of quantification. From these analyses, we selected the

Anal. Chim. Acta 664 (2010) 190–198

104

“blank” samples with the lowest levels of PBDEs, in such a way that PBDEs concentrations

in the four blanks were lower than the limits of quantification.

- Precision. The precision, expressed as the repeatability of the method, was

determined in terms of relative standard deviation (RSD, %) from recovery experiments at

each fortification level.

- Limit of quantification (LOQ) objective. It was established, from the quantification

ion, as the lowest concentration that was validated in spiked samples following the overall

procedure with satisfactory recovery (between 80% and 120%) and precision (RSD < 20%)

(0.25 μg kg−1 fish fillet and fish feed, 1 μg kg−1 fish oil and linseed oil. For BDE 209 these

values were 1.25 and 5 μg kg−1, respectively).

- Limit of detection (LOD). LOD of the method was statistically estimated, from the

quantification ion, as the analyte giving a peak signal of three times the background noise

from the sample chromatograms at the lowest fortification level tested. Instrumental LODs

were estimated from the lowest standard in solvent injected (e.g. 0.5 μg L−1, except for BDE

209 which was 2.5 μg L−1).

- Confirmation criteria: The Q/q ratio, defined as the intensity ratio between the

quantification ion (Q) and the confirmation ion (q), was used to confirm the identity of the

compounds detected in samples. Confirmation criteria were based on the European

Commission Decision (2002/657/CE) [27]. Briefly, to confirm a finding as an actual positive,

a maximum ratio tolerance ±20% was accepted when the relative intensity of the

confirmative transition was >50% as regards the quantitative one (Q/q ratio 1-2). For higher

Q/q ratios, the tolerances increased. Thus, maximum deviations of ±25% (relative intensity

20-50%, Q/q ratio 2-5), ±30% (relative intensity 10-20%, Q/q ratio 5-10) and ±50% (relative

intensity ≤ 10%, Q/q ratio > 10) were accepted. This criterion was originally defined on

measures to monitor certain substances and residues thereof in live animals and animal

products, and it is being increasingly used in other fields like environmental and biological

samples analysis [28-31]. Obviously, the agreement in the retention time in sample and

reference standard was also required to confirm a positive finding. The theoretical average

Q/q ratios for each compound were obtained as the mean value calculated from seven

standard solutions injected twice.


105

3. Results and discussion

3.1. Comparison between NPLC and SPE clean-up procedures

Firstly, the linearity of the method was evaluated with standard solutions using

responses relative to IS. The response was linear along the whole concentration range

tested, i.e. 0.5-50 μg L−1. For BDE 209, linearity was satisfactory between 2.5 μg L−1 and 250

μg L−1.

As a consequence of the high lipid content and complexity of the matrices studied,

first a clean-up of the extracts by means of acid digestion using conc. sulphuric acid was

necessary, which allowed us to remove around 90% of lipids. In spite of this, an amount of

lipids and may be other interfering compounds still remained in extracts making a second

clean-up step necessary to improve sensitivity and selectivity. The application of NPLC or

SPE was found to be an efficient way to reduce interferences, and allowed us to achieve the

low LODs pursued in this research.

NPLC and SPE (Florisil and silica) were tested in order to know the improvements of

the method in terms of LODs, LOQs, recoveries and matrix effects in the four sample

matrices fortified at 0.25 μg kg−1 (fish fillet and fish feed) or at 1 μg kg−1 (fish oil and linseed

oil). BDE 209 fortification level was 1.25 μg kg−1 and 5 μg kg−1, respectively.

All sample extracts were treated with sulphuric acid before applying the SPE or NPLC

clean-up. Recoveries obtained corresponded to the whole analytical procedure, including

extraction, sulphuric acid treatment and SPE/NPLC clean-up, and GC-(NCI)MS

measurement. The sample volume injected in the SPE/NPLC system was in all experiments

1 mL. The sample amount that can be loaded in SPE is normally higher than in NPLC. For

this reason, we performed several experiments to optimize the sample volume injected in

the NPLC system in order to avoid overloading. Finally, 1 mL was injected (notice that

sample extract was previously treated by sulphuric acid).

Although the three clean-up techniques were optimized post acid treatment, it was

found that SPE was more efficient as NPLC due to the SPE cartridges allowed the

purification of several sample extracts at the same time while, in the case of NPLC, only one

sample extract is introduced in the system. Moreover, more volume of solvents is in use in

Anal. Chim. Acta 664 (2010) 190–198

106

NPLC in 26 min. of automatic purification program [14] in comparison with SPE-Florisil or

SPE-silica.

Recoveries of PBDEs using SPE clean-up with Florisil (Table 2) were typically between

80% and 120%, and mostly better than those obtained after SPE-silica or NPLC clean-up. In

some particular cases, recoveries slightly higher than 120% were achieved suggesting a

poor correction when using DDE-D8 as surrogate, specially after silica SPE and NPLC clean-

up. Working with Florisil, only one exception was observed; BDE 154 in fish oil at 1 μg kg−1

spiked level (recovery 133%). RSDs were lower than 15% with the exception of the lowest

level (0.25 μg kg−1) in fish fillet, where values around 25% were obtained for several

compounds. In general, RSDs were also better compared with SPE-silica and NPLC.

Similar LODs were obtained with the three clean-up approaches, except for BDE 209

where the lowest LODs were reached in all matrices using Florisil SPE cartridges. Fig. 1

shows a comparison between BDE 209 chromatograms obtained when applying different

clean-up procedures to fish fillet, fish feed, linseed and fish oil samples at the lowest

fortified level (1.25 μg kg−1 for fish fillet and fish feed, 5 μg kg−1 for oils). As expected, oil

matrices presented more interferences. LODs after applying sulphuric acid treatment and

Florisil SPE clean-up are shown in Table 2. These values ranged for the eleven selected

PBDEs (in μg kg−1) from 0.03 to 0.10 (fish fillet), 0.03 to 0.19 (fish feed), 0.06 to 0.25 (fish oil)

and 0.05 to 0.20 (linseed oil). LODs for BDE 209 varied from 0.70 μg kg−1 to 0.90 μg kg−1.

Instrumental LODs were estimated from the lowest standard injected and were in the range

of 0.03-0.16 μg L−1, except for BDE 209 that was 0.52 μg L−1.

The analytical methodology developed presents high selectivity and sensitivity as

consequence of the efficient clean-up procedure applied, and the instrumental

determination by GC-(NCI)MS, reaching low LODs for all PBDEs selected in the four matrices

studied.


107

Table 2. Recoveries (%) of the overall analytical procedure for GC-(NCI)MS analysis of PBDEs in fish fillet, fish feed, fish oil and linseed oil (n=3, at each

fortification level) when using SPE-Florisil clean-up. Limits of detection.

Recoveries (%) LOD (µg kg -1

)

0.25µg kg -1

a 2.5µg kg

-1 b 1µg kg

-1 c 10µg kg

-1 d

BDE

Fish

Fillet

Fish

Feed

Fish

Fillet

Fish

Feed Fish Oil

Linseed

Oil Fish Oil

Linseed

Oil

Fish

Fillet

Fish

Feed Fish Oil

Linseed

Oil

28 107(28) 92(6) 104(2) 99(11) 87(3) 79(11) 86(24) 108(9) 0.04 0.04 0.15 0.09

71 99(25) 95(16) 102(4) 103(15) 87(8) 86(13) 89(16) 87(23) 0.04 0.10 0.07 0.05

47 98(7) 80(4) 102(10) 98(9) 92(14) 89(19) 96(7) 108(3) 0.10 0.12 0.24 0.16

66 107(19) 103(3) 98(3) 103(2) 115(9) 105(7) 93(12) 112(5) 0.10 0.19 0.25 0.20

100 115(25) 104(6) 101(9) 116(8) 107(14) 102(18) 104(10) 119(3) 0.03 0.12 0.10 0.05

99 98(4) 119(5) 103(9) 114(10) 99(7) 119(14) 104(8) 117(5) 0.07 0.07 0.08 0.12

85 106(10) 115(4) 95(5) 119(8) 103(10) 98(16) 101(5) 116(12) 0.06 0.07 0.13 0.12

154 110(5) 117(9) 101(11) 126(6) 133(11) 101(10) 98(6) 111(5) 0.05 0.03 0.08 0.08

153 95(26) 108(13) 100(15) 107(8) 108(10) 108(3) 107(7) 118(15) 0.06 0.05 0.11 0.09

138 100(25) 102(6) 102(16) 132(6) 113(9) 111(3) 102(11) 119(1) 0.08 0.05 0.15 0.16

183 100(21) 110(5) 75(12) 100(8) 103(10) 109(2) 81(9) 99(7) 0.04 0.05 0.06 0.07

209 110(16) 84(15) 80(13) 116(8) 104(12) 106(9) 84(8) 109(8) 0.70 0.90 0.90 0.85 a 1.25 µg kg

-1 for BDE 209,

b 12.5 µg kg

-1 for BDE 209.

c 5 µg kg

-1 for BDE 209,

d 50 µg kg

-1 for BDE 209.

RSD (%) shown in brackets.

Anal. Chim. Acta 664 (2010) 190–198

108

Fig. 1. GC-(NCI)MS chromatograms of BDE 209 for fish fillet, fish feed, linseed and fish oils spiked at

the lowest fortification level (1.25 μg kg−1

for fish fillet and fish feed, 5 μg kg−1

for oils). Comparison

of different clean-up procedures: SPE-Florisil (top), SPE-silica (medium), and NPLC (bottom) applied

after sulphuric acid treatment.

Fig. 2 and Fig. 3 show illustrative GC-(NCI)MS chromatograms at the lowest level

validated for the sample matrices studied after applying the overall recommended

analytical procedure. Fig. 2 shows chromatograms for tri-BDE (28) and tetra-BDEs (47, 66,

71) and Fig. 3 shows chromatograms for penta-BDEs (85, 99, 100), hexa-BDEs (138, 153,

154), hepta-BDE (183) and deca-BDE (209). As it can be seen, satisfactory sensitivity was

obtained at the LOQ level when using the quantification ion (Q) and the first confirmation

ion (q1), while the lower abundance of the second confirmation ion (q2) led to a notable

decrease in sensitivity. In spite of this, q2 was also acquired to help in the confirmation of

the identity of positive findings.


109

Fig. 2. GC-(NCI)MS illustrative chromatograms at 0.25 μg kg

−1 level for fish fillet (top) and fish feed

(bottom) following the overall recommended procedure. “Q” quantification ion, “q1 and q2”

confirmation ions. Q/q were in agreement with the criteria established. Numbers on top of the peak

are retention time (top) and peak area (bottom).

Fig. 3. GC-(NCI)MS illustrative chromatograms at 1 μg kg

−1 level (5 μg kg

−1 for BDE 209) for linseed oil

(top) and fish oil (bottom) following the overall recommended procedure. “Q” quantification ion, “q1

and q2” confirmation ion. Q/q were in agreement with the criteria established. Numbers on top of

the peak are retention time (top) and peak area (bottom).

Anal. Chim. Acta 664 (2010) 190–198

110

3.2. GC-(NCI)MS conditions

GC coupled to mass spectrometry is, in most cases, the preferred technique for the

analysis of organic pollutants in environmental samples [3,32]. In addition, the QqQ mass

analyzer working in tandem MS/MS mode provides high selectivity and sensitivity in the

analysis of complex environmental samples, which is an evident advantage in

environmental analysis. The use of two stages of mass analysis in MS/MS systems based on

QqQ analyzers offers the possibility of applying selected reaction monitoring, one of the

most selective and sensitive approaches. In spite of this, there are still some compounds

that cannot be determined with enough sensitivity with this technique. Thus, the

development of MS/MS method in NCI mode for PBDEs was not feasible because of the

poor MS spectra, which only showed the clusters from the fragments [Br]− and [HBr2]− [6].

The molecular cluster was not observed or only constituted a minor peak. Therefore, the

unique transition reasonable for the majority of compounds would have been the

fragmentation of [HBr2]− to give a bromine atom, with low sensitivity and poor selectivity.

Consequently, a SIR method was optimized by monitoring the three most intense peaks of

the mass spectra, which corresponded to m/z 79 ([79Br]−), m/z 81 ([81Br]−), m/z 161

([H79Br81Br]−) and m/z 487 ([C6O79Br3

81Br2]−) for BDE 209.

Chromatographic conditions for PBDEs in GC-MS were optimized by injecting

standard solutions in solvent. We studied the temperature program in order to achieve

satisfactory resolution and peak shape for the IS and 12 PBDEs, including the BDE 209. The

deca-BDE 209 should receive special attention because of its sensitivity for the higher

susceptibility for degradation in the GC system [33]. Due to its unique chemical and physical

properties, including high molecular weight, accurate determination of BDE 209 still

remains a challenge, especially at low analyte levels. The different oven programs tested in

our work indicated that reducing the column residence time of BDE 209, by using a short 15

m column, in combination with fast ramp rates, improved the peak sensitivity. Increasing

oven temperature up to 340 °C and setting the injection temperature to 260 °C, to ensure

complete vaporization, resulted in a consistent and more sensitive response, notably


111

reducing the BDE 209 retention time. Therefore, a short and thin capillary column was used

to prevent degradation of PBDEs and to improve accuracy and precision of the analysis. The

overall procedure was tested for the analytical characteristics shown in the validation

section, obtaining satisfactory results. Due to its special characteristics, BDE 209 is not

commonly included in multi-analyte PBDEs methods [34, 35].

The coelution of PBDE congeners under certain chromatographic conditions has been

reported. Thus, BDE 49 and BDE 71 are close-eluting on 100% Dimethylpolysiloxane GC

capillary column and might coelute in our DB-1HT column. Covaci et al. [13] have

performed a detailed study of PBDEs chromatographic behavior using different GC capillary

columns, but they reported no coelution of the two mentioned congeners. We did not test

this fact as BDE 49 was not studied in our work.

NCI ionization in SIR acquisition mode was applied to monitoring bromine ions:

[79Br]−, [81Br]− and secondary fragments such [H79Br81Br]− and [C6O79Br3

81Br2]−. GC-(NCI)MS

was proven to notably increase the selectivity, sensitivity and accuracy for PBDEs

determination, specially for BDE 209 [13] and [16]. The [Br]− mass fragment was the most

intense ion, while the molecular ion was normally missing. [H79Br81Br]− and [C6O79Br3

81Br2]−

fragments were less abundant than [Br]−. Thus, [Br]− (m/z = 79) was selected for

quantification due to the higher sensitivity reached, and [Br]− (m/z = 81) was selected as

confirmation ion. [H79Br81Br]− and [C6O79Br3

81Br2]− were selected for a more reliable

confirmation of PBDEs in samples, but they presented some difficulties due to the lower

abundance compared with [Br]− ions. Due to the notable differences in abundances, we

obtained Q/q ratios around one for the relation with 79/81 (Q/q1), but much higher Q/q

ratios for 79/161 and 79/487 (Q/q2). Fig. 2, Fig. 3 and Fig. 4 illustrate in fact the limitations

for confirmation when using Q/q2 ratio in samples at low concentration levels. In spite of

this, we decided to acquire m/z = 161 and 487 although in the most of cases the Q/q2 ratios

were higher than 20. Using the second confirmation ion was useful when the analyte was

present at higher concentrations. Agreement in Q/q ratios between samples and reference

standards is required to confirm positive findings [27]. The agreement in the retention time

of a compound in the sample and in reference standard is also required to confirm a

positive result.

Anal. Chim. Acta 664 (2010) 190–198

112

3.3. Application to samples from bioaccumulation experiments

The overall optimized procedure was applied to the analysis (in triplicate) of 20 fish

fillet samples, 11 fish feed, 1 fish oil, 1 linseed oil, 1 rapeseed oil and 1 palm oil used in

aquaculture experiments. The results obtained are shown in Table 3. Most PBDE congeners

were below the limit of detection and only a few samples presented positive findings in this

study. The most frequently detected congeners were PBDE 28 and PBDE 47. Fig. 4 reveals

positive findings of tetra, penta- and hexa-PBDEs in fish fillets. Concentration levels found in

fish fillets ranged from 0.25 μg kg−1 to 4.7 μg kg−1. Some authors have also reported low

PBDE levels in wild aquatic species (mud carp and crucian carp), with individual values from

1.3 μg kg−1 to 400 μg kg−1 [5]. Shaw et al. [18] reported mean concentrations of ∑PBDEs in

farmed salmon samples from Maine (0.95 μg kg−1 ww, 7.3 μg kg−1 lw) and eastern Canada

(0.85 μg kg−1 ww, 6.3 μg kg−1 lw) lower than data reported in our paper for fish fillet.

Fig. 4. Positive findings of PBDEs in samples of fish fillet, fish feed. Q/q were in agreement with the

criteria established.


113

PBDEs in fish feed ranged from 0.5 μg kg−1 to 2.8 μg kg−1, except for one sample that

contained BDE 209 at a 15.5 μg kg−1 (Fig. 4). This high value corresponds to the only

commercial fish feed analyzed, as the remaining samples were home-made for the

AQUAMAX project after a careful selection of the ingredients. The high BDE 209 level was

confirmed by additional analysis, concluding that there was a remarkable contamination in

this sample. Some authors reported low contamination by PBDEs in aquatic species

compared with the samples in our work [18] and [36]. Dietary intake studies of lower PBDEs

have shown that fish and animal products are important vectors of human exposure, but

data on higher brominated BDEs are scarce [37].

Table 3. PBDEs detected in aquaculture samples (n=3).

Fish Fillet

Fish Feed

BDE

Congener Number of positives

conc. range

(µg kg -1

) Number of positives

conc. range

(µg kg -1

)

28 7 0.25-0.5 4 0.25-0.3

71 3 0.25-0.5 4 0.25-0.8

47 14 0.25-4.7 4 0.25-2.8

66 1 0.6 3 0.25-0.3

100 2 0.4-0.7 4 0.25-0.7

99 2 0.6-0.9 4 0.25-0.9

85 2 0.4-0.7 4 0.25-0.6

154 2 0.4-0.6 4 0.25-0.6

153 2 0.4-0.6 4 0.4-0.9

138 1 0.7 2 0.25-0.3

183 1 0.6 2 0.25-0.3

209 - 1 15.5

Total number of samples analyzed: 20 fish fillet, 11 fish feed. Although some PBDEs were detected in

the oil samples, their levels were below the limit of quantification.

We did not find PBDEs in the oil samples analyzed at levels higher than the LOQ

objective (1 μg kg−1; 5 μg kg−1 BDE 209). This fact agrees with data reported by Domingo et

al. [9], where total load of PBDEs from marine species, oils and fats and other components

from the normal diet was low; e.g. the highest PBDE levels found in oils and fats was 0.59

μg kg−1.

Anal. Chim. Acta 664 (2010) 190–198

114

4. Conclusions

An efficient and sensitive analytical methodology for the determination of PBDEs in

fat aquaculture matrices has been developed using an efficient clean-up step combining

acid digestion and SPE-Florisil procedure previous to the injection of the extracts in GC-MS

system operating in NCI ionization and SIR mode acquisition. The procedure applied

allowed to reach LODs as low as 0.03-0.06 μg kg−1 in samples. The most difficult task in this

work was to determine the BDE 209, as it required special chromatographic conditions to

reach good peak shape and satisfactory sensitivity. The acquisition of three m/z ions in SIR

mode offers a reliable confirmation of the identity of the PBDEs detected in samples. BDE

28 and 47 were the most commonly found in fish fillet and fish feed. Individual

concentrations were always below 5 μg kg−1, except for BDE 209 in one fish feed sample

that reached up to 15.5 μg kg−1.

Acknowledgements

The authors acknowledge the financial support of the European Union Project:

Sustainable Aquafeeds to Maximise the Health Benefits of Farmed Fish for Consumers

(AQUAMAX). Contract number: 016249-2.

References

[1] S.D. Richardson, Anal. Chem. 80 (2008), pp. 4373-4402.

[2] A. Binelli, S.K. Sarkar, M. Chatterjee, C. Riva, M. Parolini, B.d. Bhattacharya, A.K.

Bhattacharya and K.K. Satpathy, Mar. Pollut. Bull. 54 (2007), pp. 1220-1229.

[3] F. Hernández, T. Portolés, E. Pitarch and F.J. López, Anal. Chem. 79 (2007), pp. 9494-

9504.

[4] S.D. Shaw and K. Kannan, Rev. Environ. Health 24 (2009), pp. 157-229.

[5] J. Wu, X. Luo, Y. Zhang, Y. Luo and S. Chen, Environ. Int. 34 (2008), pp. 1109-1113.


115

[6] C.M. Medina, E. Pitarch, F.J. López, C. Vázquez and F. Hernández, Anal. Bioanal. Chem.

390 (2008), pp. 1343-1354.

[7] C.M. Medina, E. Pitarch, T. Portolés, F.J. López and F. Hernández, J. Sep. Sci. 32 (2009),

pp. 2090-2102.

[8] B.C. Kelly, M.G. Ikonomou, J.D. Blair and F.A.P.C. Gobas, Sci. Total Environ. 401 (2008),

pp. 60-78.

[9] J.L. Domingo, R. Mara-Cid, V. Castell and J.M. Llobet, Toxicology 248 (2008), pp. 25-32.

[10] G.T. Yogui and J.L. Sericano, Environ. Int. 35 (2008), pp. 655-666.

[11] S. Voorspoels, A. Covaci and P. Schepens, Environ. Sci. Technol. 37 (2003), pp. 4348-

4357.

[12] M.N. Jacobs, A. Covaci, A. Gheorghe and P. Schepens, J. Agric. Food Chem. 52 (2004),

pp. 1780- 1788.

[13] A. Covaci, S. Voorspoels, L. Ramos, H. Neels and R. Blust, J. Chromatogr. A 1153 (2007),

pp. 145-171.

[14] R. Serrano, M. Barreda, E. Pitarch and F. Hernández, J. Sep. Sci 26 (2003), pp. 75-86.

[15] R. Cariou, J.-P. Antignac and L. Debrauwer, J. Chromatogr. Sci 44 (2006), pp. 489-497.

[16] H.M. Stapleton, Anal. Bioanal. Chem 386 (2006), pp. 807-817.

[17] S. Bayen, H.K. Lee and J.P. Obbard, J. Chromatogr. A 1035 (2004), pp. 291-294.

[18] S.D. Shaw, M.L. Berger, D. Brenner, D.O. Carpenter, L. Tao, C. Hong and K. Kannan,

Chemosphere 71 (2008), pp. 1422-1431.

[19] A.R. Fontana, M.F. Silva, L.D. Martínez, R.G. Wuilloud and J.C. Altamirano, J.

Chromatogr. A 1216 (2009), pp. 4339-4346.

[20] J. Björklund, P. Tollbäch and C. Östman, J. Mass Spectrom. 38 (2003), pp. 394-400.

[21] A. Covaci, J. De Boer, J.J. Ryan, S. Voorspoels and P. Schepens, Anal. Chem. 74 (2002),

pp. 790-798.

[22] A.M. Carro, R.A. Lorenzo, F. Fernández, R. Phan-Tan-Luu and R. Cela, Anal. Bioanal.

Chem. 388 (2007), pp. 1021-1029.

[23] L. Roosens, A.C. Dirtu, G. Goemans, C. Belpaire, A. Gheorghe, H. Neels, R. Blust and A.

Covaci, Environ. Int. 34 (2008), pp. 976-983.

Anal. Chim. Acta 664 (2010) 190–198

116

[24] J. Hajšlová, J. Pulkrabová, J. Poustka, T. Căjka and T. Randák, Chemosphere 69 (2007),

pp. 1195-1203.

[25] M. Frederiksen, K. Vorkamp, M. Thomsen and L.E. Knudsen, Int. J. Hyg. Environ. Health

212 (2009), pp. 109-134.

[26] N.M. Fajar, A.M. Carro, R.A. Lorenzo, F. Fernandez and R. Cela, Food Addit. Contam. A

25 (2008), pp. 1015-1023.

[27] Commission Decision 2002/657/EC implementing Council Directive 96/23/EC

concerning the performance of analytical methods and the interpretation of results

(eur-lex.europa.eu/pri/en/oj/dat/2002/l_221/l_22120020817en00080036.pdf).

[28] F. Hernández, J.V. Sancho and O.J. Pozo, Anal. Bioanal. Chem. 382 (2005), pp. 934-946.

[29] I. Ferrer, E.T. Furlong and E.M. Thurman In: I. Ferrer and E.M. Thurman, Editors, Liquid

Chromatography Mass Spectrometry/Mass Spectrometry, MS/MS and Time-of-Flight

MS: Analysis of Emerging Contaminants, American Chemical Society Symposium Series

850, Oxford University Press, New York (2003), pp. 376-393.

[30] O.J. Pozo, J.V. Sancho, M. Ibáñez, F. Hernández and W.M.A. Niessen, Trends Anal.

Chem. 25 (2006), pp. 1030-1042.

[31] M.W.F. Nielen, M.C. van Engelen, R. Zuiderent and R. Ramaker, Anal. Chim. Acta 586

(2007), pp. 122-129.

[32] J. Nácher-Mestre, R. Serrano, T. Portolés, F. Hernández, L. Benedito-Palos and J. Pérez-

Sánchez, Rapid Commun. Mass Spectrom. 23 (2009), pp. 2075-2086.

[33] A. Covaci, S. Voorspoels and J. De Boer, Environ. Int. 29 (2003), pp. 735-756.

[34] Y.P. Liu, J.G. Li, Y.F. Zhao, Y.N. Wu and L.Y. Zhu, Food Addit. Contam. A 26 (2009), pp.

1180-1184.

[35] S. Losada, F.J. Santos and M.T. Galcerán, Talanta 80 (2009), pp. 839-845.

[36] D. Hayward, J. Wong and A.J. Krynitsky, Environ. Res. 103 (2007), pp. 46-54.

[37] A. Kierkegaard, L. Asplund, C.A. De Wit, M.S. McLachlan, G.O. Thomas, A.J. Sweetman

and K.C. Jones, Environ. Sci. Technol. 41 (2007), pp. 417-423.


117

SUPPORTING DATA

Figure 1. SIR chromatogram for the IS (DDE-D8, m/z 289) and 12 PBDEs (m/z 79) spiked at the lowest level in fish feed (0.25µg/kg, and

1.25µg/kg for BDE 209).

pienso florisil 3 F5/25

8.00 9.00 10.00 11.00 12.00 13.00 14.00 15.00 16.00

%

0

10012.53852

9.82779

9.55350

7.85134

11.30612

10.11179

11.69613

12.20394

12.90585

13.81373

13.29215

16.4660

7.3710

Anal. Chim. Acta 664 (2010) 190–198

118

SUPPORTING DATA

Table1. Recoveries of the overall analytical procedure for spiked samples when using different SPE or NPLC clean-up.

Florisil Silica NPLC

0.25 µg/kg 1 µg/kg 0.25 µg/kg 1 µg/kg 0.25 µg/kg 1 µg/kg

BDE Filete Pienso Aceite

pescado

Aceite

lino Filete Pienso

Aceite

pescado

Aceite

lino Filete Pienso

Aceite

pescado

Aceite

lino

28 107(28) 92(6) 87(3) 79(11) 118(23) 115(16) 63(9) 73(12) 95(15) 82(4) 63(20) 102(15)

71 99(25) 95(16) 87(8) 86(13) 97(24) 145(9) 68(21) 78(15) 100(23) 105(11) 60(14) 102(12)

47 98(7) 80(4) 92(14) 89(19) 60(17) 129(18) 60(11) 78(23) 56(29) 64(28) 55(14) 116(13)

66 107(19) 103(3) 115(9) 105(7) 112(22) 122(10) 89(4) 98(7) 107(26) 73(8) 94(9) 135(16)

100 115(25) 104(6) 107(14) 102(18) 87(16) 137(11) 73(16) 93(20) 95(21) 107(2) 68(10) 126(14)

99 98(4) 119(5) 99(7) 119(14) 96(22) 130(25) 108(16) 118(21) 115(17) 105(14) 93(10) 172(17)

85 106(10) 115(4) 103(10) 98(16) 91(27) 175(13) 94(12) 90(14) 95(24) 110(5) 80(11) 104(9)

154 110(5) 117(9) 133(11) 101(10) 75(24) 184(9) 104(17) 108(20) 102(14) 110(5) 85(21) 144(15)

153 95(26) 108(13) 108(10) 108(3) 84(8) 170(8) 84(5) 110(22) 96(11) 117(3) 75(12) 155(15)

138 100(25) 102(6) 113(9) 111(3) 97(11) 145(6) 99(14) 113(24) 110(6) 93(3) 80(13) 140(14)

183 100(21) 110(5) 103(10) 109(2) 95(7) 165(11) 94(15) 113(20) 95(26) 120(4) 80(26) 149(25)

209 110(16) 84(15) 104(12) 106(9) 134(24) 159(25) 94(14) 118(17) 125(7) 73(28) 63(21) 107(16)


119

SUPPORTING DATA

Table 2. LODs obtained in the four sample matrices analyzed using different clean-up procedures.

Florisil Silica NPLC

Compound Fish

Fillet

Fish

Feed

Fish

Oil

Linseed

Oil

Fish

Fillet

Fish

Feed

Fish

Oil

Linseed

Oil

Fish

Fillet

Fish

Feed

Fish

Oil

Linseed

Oil

28 0.04 0.04 0.15 0.09 0.05 0.07 0.13 0.16 0.08 0.04 0.16 0.12

71 0.04 0.10 0.07 0.05 0.04 0.20 0.08 0.05 0.07 0.09 0.06 0.05

47 0.10 0.12 0.24 0.16 0.06 0.16 0.19 0.21 0.15 0.09 0.16 0.27

66 0.10 0.19 0.25 0.20 0.10 0.21 0.19 0.24 0.30 0.06 0.28 0.28

100 0.03 0.12 0.10 0.05 0.05 0.16 0.08 0.05 0.06 0.10 0.06 0.05

99 0.07 0.07 0.08 0.12 0.05 0.08 0.14 0.11 0.10 0.09 0.08 0.13

85 0.06 0.07 0.13 0.12 0.06 0.12 0.13 0.12 0.20 0.07 0.12 0.11

154 0.05 0.03 0.08 0.08 0.04 0.07 0.05 0.05 0.04 0.10 0.04 0.07

153 0.06 0.05 0.11 0.09 0.07 0.07 0.12 0.08 0.09 0.07 0.07 0.12

138 0.08 0.05 0.15 0.16 0.08 0.08 0.16 0.13 0.10 0.04 0.15 0.19

183 0.04 0.05 0.06 0.07 0.06 0.06 0.05 0.05 0.05 0.06 0.06 0.09

209 0.70 0.90 0.90 0.85 3.30 1.23 2.90 2.30 2.41 2.23 2.15 2.50


121

4.3.3. Discusión de resultados.

Debido al gran uso de los PBDEs durante años y aún utilizados en gran cantidad de

materiales en servicio hoy en día, su presencia en diferentes matrices ambientales ha sido

puesta de manifiesto por varios autores (Kierkegaard et al., 2009). A pesar de que existen

hasta 209 congéneres de PBDEs, no todos son frecuentemente encontrados en muestras

ambientales. Generalmente los congéneres mayoritariamente identificados y cuantificados

son el BDE-47 (2,2',4,4'-tetrabromodiphenyl ether), el BDE-99 (2,2',4,4',5-

pentabromodiphenyl ether), BDE-100 (2,2',4,4',6-pentabromodiphenyl ether) y en muchos

casos BDE-153 (2,2',4,4',5,5'-hexabromodiphenyl ether), BDE-154 (2,2',4,4',5,6'-

hexabromodiphenyl ether) y el BDE (209) (decabromodiphenyl ether) (Lacorte et al., 2010;

Zhang et al., 2010; Boon et al., 2002). Aunque son los PBDEs de mayor bromación los que

presentan mayor tendencia a la bioacumulación y biomagnificación (BDEs con kow>4), los

PBDEs de menor bromación (tetra-BDEs y penta-BDEs) suelen presentar niveles altos en

muestras ambientales y por eso es también importante su análisis. La lista de PBDEs

propuesta en nuestra metología ha sido elaborada en virtud de la presencia de estos PBDEs

en las matrices seleccionadas procedentes de la acuicultura marina. Así pues, PBDEs con

diferente grado de bromación han sido seleccionados desde el tri-BDE-28 hasta el deca-

BDE-209. La metodología desarrollada permite la determinación de un mayor número de

congéneres en comparación con algunas metodologías recientes en las que el estudio de

PBDEs se centra únicamente en un grupo reducido de éstos (Fontana et al., 2009; Liu et al.,

2009; Montes et al., 2009; Chen et al., 2010).

• Estudio de la etapa de purificación.

Como se ha indicado con anterioridad, gran parte de los interferentes presentes en

las muestras de estudio, principalmente lípidos, dificultan el análisis de los compuestos

orgánicos y concretamente de los PBDEs. La elección de un procedimiento de purificación

adecuado que mejore la selectividad y sensibilidad del método analítico para la

determinación de estos contaminantes en las muestras de estudio será un factor decisivo.


122

Es necesario un sistema de purificación eficaz que elimine las sustancias interferentes en la

determinación cromatográfica, evitando daños en la columna y la contaminación de los

detectores de masas. En este sentido, el tratamiento con ácido resulta una técnica que

permite eliminar prácticamente la gran mayoría de los lípidos presentes en la matriz (Covaci

et al., 2003). Serrano et al. (2003) ya propuso una metodología similar a la del presente

estudio para eliminar la fracción lipídica de muestras procedentes de la acuicultura. Más del

90% de las grasas eran eliminadas mediante un tratamiento con H2SO4 (Serrano et al.,

2003). A pesar de este primer tratamiento, la gran cantidad de lípidos presentes en las

muestras inicialmente hace necesaria una etapa adicional de purificación. En el presente

estudio, se han ensayado tres procedimientos de purificación diferentes, comparándose la

eficacia de cada uno en la eliminación de interferentes de la matriz.

En concreto se ensayaron un sistema automatizado de purificación mediante NPLC,

además de la SPE mediante cartuchos con fase estacionaria de Florisil (SPE-Florisil) y de

silica (SPE-silica). Los tres procedimientos se aplicaron sobre extractos de las muestras de

estudio después de su extracción y posterior tratamiento con ácido. Se realizaron tres

réplicas de cada muestra a un nivel de fortificación de 0.25 µg/kg (filete de dorada y pienso)

y de 1 µg/kg (aceite de pescado y aceite de lino). Estos extractos purificados se analizaron

mediante GC-(NCI)MS y se compararon los resultados atendiendo a los LOD, LOQ,

recuperación absoluta y efecto matriz.

Los resultados mostraron que la etapa de purificación mediante SPE-Florisil resultó

más eficaz que las otras técnicas por conseguir una mayor eficacia de purificación. Además

el proceso consume menos tiempo y menor gasto de disolventes. Igualmente, los LODs

globales obtenidos para las diferentes etapas fueron óptimos en todos los casos excepto

para el BDE 209 en donde los mejores LODs fueron obtenidos con SPE-Florisil (ver Tabla 2,

Supporting Data, Artículo científico 2). Las recuperaciones absolutas obtenidas para las

diferentes muestras fortificadas al nivel bajo, mediante los tres procedimientos de

purificación, pueden verse en la Tabla 1 (Supporting Data, Artículo científico 2). Como se

puede observar, estos valores son mejores mediante el procedimiento de SPE-Florisil en

comparación con los obtenidos utilizando SPE-silica y NPLC que, para el caso de algunos

PBDEs, presentaban valores muy elevados.


123

La aplicación de una segunda etapa de purificación posterior al tratamiento con

H2SO4 resultó muy eficaz. Se redujeron de una manera notable los interferentes, lo que

permitió alcanzar LODs a los niveles deseados.

• Optimización GC-MS.

Respecto a la determinación instrumental, se seleccionó el modo de ionización NCI

en lugar de EI, puesto que la sensibilidad en modo CI era mayor. Este hecho fue contrastado

con bibliografía disponible (Eljarrat et al., 2002; Covaci et al., 2003). El desarrollo de una

metodología con fragmentación en MS/MS en modo NCI fue rechazada puesto que el

espectro de masas resultante era muy poco significativo por presentar únicamente los

fragmentos [Br]- y [HBr2]

-. La única transición posible para la mayoría de PBDEs resultó ser

la fragmentación de [HBr2]- para observar [Br]

-. Adquiriendo en modo SIR los iones [Br]

- y

[HBr2]- se obtuvo mayor sensibilidad que para la fragmentación en MS/MS por lo que se

decidió trabajar en modo SIR adquiriendo tres iones por compuesto (ver Tabla 1, Artículo

científico 2). La adquisición de tres iones en modo SIR ofreció una buena confirmación de la

identidad de los PBDEs en las muestras. Cabe destacar que los iones m/z1=79, m/z2=81

procedentes del fragmento [Br]- eran los más sensibles en comparación a m/z3=161 del

fragmento [HBr2]-.

• Determinación del PBDE 209.

El PBDE 209 es el difenil éter polibromado de mayor bromación lo que hace que

tenga unas características físico-químicas específicas como son su pobre volatilidad y la

posibilidad de degradación en los inyectores y en las columnas del sistema cromatográfico.

Como consecuencia, este compuesto presenta más dificultades que ningún otro PBDE en su

análisis mediante GC-MS (Stapleton et al., 2006; Covaci et al., 2007). Por este motivo no se

ha incluido frecuentemente al PBDE 209 en metodologías publicadas (Borghesi et al., 2009;

Liu X. et al., 2009; Bianco et al., 2010) o se obtienen recuperaciones no satisfactorias en

algunos casos (Labadie et al., 2010a). El PBDE 209 es también el compuesto que presenta


124

una menor sensibilidad por lo que, comparándolo con los demás PBDEs de la lista, puede

llegar a ser incluso 5 veces menos sensible (Figura 4.3). Esto hizo necesario el estudio de la

validación de este compuesto a un nivel 5 veces mas concentrado que para el resto de los

PBDEs. Por su degradabilidad en el sistema cromatográfico es conveniente disminuir, en la

medida de lo posible, el tiempo de residencia en el sistema. Aplicando una rampa de

temperaturas rápida y alcanzando los 340°C se consiguió, mediante una columna DB-1HT

de 15 m, que el tiempo de elución y el tiempo de residencia fuera reducido. Otros autores

también han sugerido la necesidad de utilizar columnas cortas para la determinación de los

PBDEs con mayor índice de bromación, obteniendo resultados satisfactorios (Lacorte et al.,

2010; Labadie et al., 2010b).

Time8.00 9.00 10.00 11.00 12.00 13.00 14.00 15.00 16.00 17.00

%

0

100

8.00 9.00 10.00 11.00 12.00 13.00 14.00 15.00 16.00 17.00

%

0

100

SIR of 6 Channels CI-79

1.28e4Area

9.82;951

7.87667

11.25;912

10.11444

12.48;815

11.63598

12.15377

12.87522

13.81375

13.29284

SIR of 6 Channels CI-79

1.53e5Area

9.82;12253

9.5663277.85

5012

11.306897

10.113183

12.535996

12.193775

12.904277

13.82256213.29

2048

16.00 16.50 17.00

%

0

100 16.47;25

Time16.00 16.50 17.00

%

0

100 16.46;282

A

B

Figura 4.3. Cromatogramas GC-(NCI)MS. Mezcla patrón de referencia a 0.5 µg/L (2.5 µg/L

en BDE 209) en hexano (A). Blanco de pienso fortificado a 0.25 µg/kg (1.25 µg/kg en BDE

209)(B).


125

• Validación del método analítico.

La metodología desarrollada fue validada en cuatros matrices: filete de dorada,

pienso, aceite de pescado y aceite de lino. Como el contenido lipídico en aceites era

aproximadamente cuatro veces mayor que en filetes y pienso y durante el procedimiento

los extractos de muestras de aceite iban a ser diluidos para minimizar el efecto matriz, la

valización en aceites se llevó a cabo a una concentración cuatro veces mayor: 1 μg kg−1

y 10

μg kg−1

(5 μg kg−1

y 50 μg kg−1

para BDE 209). El proceso de validación se llevó a cabo

atendiendo a los parámetros propuestos por la guía SANCO vigente (Document N°

SANCO/2007/3131).

• Aplicación de la metodología desarrollada a muestras reales.

Una vez desarrollado y validado el método analítico mediante GC-(NCI)MS, éste se

aplicó al análisis de muestras reales. Se analizaron 20 muestras de filete de dorada que

habían sido alimentadas mediante piensos con diferente composición procedentes del

proyecto “AQUAMAX”. También se analizaron 11 piensos con diferente carga lipídica y

composición de aceites vegetales y de origen animal. Por último, la metodología también se

aplicó para el análisis de cuatro aceites: aceite de pescado, aceite de lino, aceite de colza y

aceite de palma.

En general los valores de PBDEs encontrados en nuestro estudio fueron muy bajos

(0.25-4.7 µg/kg en filete, 0.5-15.5 µg/kg en piensos). En algunos casos, las muestras no

fueron cuantificadas por presentar concentraciones por debajo del LOQ propuesto (0.25

μg/kg en filete de pescado y pienso, 1 μg/kg en aceites; para el BDE 209 estos valores

fueron 1.25 y 5 μg/kg, respectivamente).

Los ingredientes de origen marino, tradicionalmente usados en la alimentación de

pescado comercial, son la fuente de contaminantes en el pescado cultivado. El uso de

materias primas sostenibles y seleccionadas con niveles bajos de PBDEs da lugar a piensos y

peces cultivados con niveles bajos de estos compuestos. Los valores propuestos en este

estudio (ver Tabla 3, Artículo científico 2) son comparables con los valores encontrados


126

por otros autores que revelan poca contaminación en peces cultivados utilizando materias

primas con baja carga de contaminantes (Berntssen et al., 2010).

4.3.4. Referencias.

Berntssen, M.H.G.; Julshamn, K.; Lundebye, A.-K. 2010. Chemical contaminants in

aquafeeds and Atlantic salmon (Salmo salar) following the use of traditional- versus

alternative feed ingredients. Chemosphere 78, 637-646.

Bianco, G., Novario, G., Anzilotta, G., Palma, A., Mangone, A., Cataldi, T.R.I. 2010.

Polybrominated diphenyl ethers (PBDEs) in Mediterranean mussels (Mytilus

galloprovincialis) from selected Apulia coastal sites evaluated by GC-HRMS. J. Mass

Spectrom. 45, 1046-1055.

Boon, J.P.; Lewis, W.E.; Tjoen-A-Choy, M.R.; Allchin, C.R.; Law, R.J.; De Boer, J.; Ten Hallers-

Tjabbes, C.C.; Zegers, B.N. 2002. Levels of polybrominated diphenyl ether (PBDE)

flame retardants in animals representing different trophic levels of the North Sea

food web. Environ. Sci. Technol. 36, 4025-4032.

Borghesi, N., Corsolini, S., Leonards, P., Brandsma, S., de Boer, J., Focardi, S. 2009.

Polybrominated diphenyl ether contamination levels in fish from the Antarctic and

the Mediterranean Sea. Chemosphere 77, 693-698.

Chen, M.Y.Y.; Tang, A.S.P.; Ho, Y.Y.; Xiao, Y. 2010. Dietary exposure of secondary school

students in Hong Kong to polybrominated diphenyl ethers from foods of animal

origin. Food Addit. Contam. 27, 521-529.

Covaci, A.; Voorspoels, S.; de Boer, J. 2003. Determination of brominated flame retardants,

with emphasis on polybrominated diphenyl ethers (PBDEs) in environmental and

human samples. Environ. Int. 29, 735-756.

Covaci, A.; Voorspoels, S.; Ramos, L.; Neels, H.; Blust, R. 2007. Recent developments in the

analysis of brominated flame retardants and brominated natural compounds. J.

Chromatogr. A 1153, 145-171.


127

Document N° SANCO/2007/3131. 2007. Method validation and quality control procedures

for pesticide residues analysis in food and feed.

Eljarrat, E.; Lacorte, S.; Barcelo, D. 2002. Optimization of congener-specific analysis of 40

polybrominated diphenyl ethers by gas chromatography/mass spectrometry. J. Mass

Spectr. 37, 76-84.

Fontana, A.R.; Silva, M.F.; Martínez, L.D.; Wuilloud, R.G.; Altamirano, J.C. 2009.

Determination of polybrominated diphenyl ethers in water and soil samples by cloud

point extraction-ultrasound-assisted back-extraction-gas chromatography-mass

spectrometry. J. Chromatogr. A 1216, 4339-4346.

Frederiksen, M.; Vorkamp, K.; Thomsen, M.; Knudsen, L.E. 2009. Human internal and

external exposure to PBDEs - A review of levels and sources. Int. J. Hyg. Envir. Heal.

212, 109-134.

Kierkegaard, A.; Sellström, U.; McLachlan, M.S. 2009. Environmental analysis of higher

brominated diphenyl ethers and decabromodiphenyl ethane. J. Chromatogr. A 1216,

364-375.

Labadie, P.; Tlili, K.; Alliot, F.; Bourges, C.; Desportes, A.; Chevreuil, M. 2010a. Development

of analytical procedures for trace-level determination of polybrominated diphenyl

ethers and tetrabromobisphenol A in river water and sediment. Anal. Bioanal. Chem.

396, 865-875.

Labadie, P.; Alliot, F.; Bourges, C.; Desportes, A.; Chevreuil, M. 2010b. Determination of

polybrominated diphenyl ethers in fish tissues by matrix solid-phase dispersion and

gas chromatography coupled to triple quadrupole mass spectrometry: Case study on

European eel (Anguilla anguilla) from Mediterranean coastal lagoons. Anal. Chim.

Acta 675, 97-105.

Lacorte, S.; Ikonomou, M.G.; Fischer, M. 2010. A comprehensive gas chromatography

coupled to high resolution mass spectrometry based method for the determination

of polybrominated diphenyl ethers and their hydroxylated and methoxylated

metabolites in environmental samples. J. Chromatogr. A 1217, 337-347.

Liu, X., Hu, J., Huang, C., Wang, H., Wang, X. 2009. Determination of polybrominated

diphenyl ethers in aquatic animal tissue using cleanup by freezing-dispersive liquid-

liquid microextraction combined with GC-MS. J. Sep. Sci. 32, 4213-4219.


128

Liu, Y.P.; Li, J.G.; Zhao, Y.F.; Wu, Y.N.; Zhu, L.Y. 2009. Rapid determination of

polybrominated diphenyl ethers (PBDEs) in fish using selective pressurized liquid

extraction (SPLE) combined with automated online gel permeation chromatography-

gas chromatography mass spectrometry (GPC-GC/MS). Food Addit. Contam. A 26,

1180-1184.

Montes, R.; Rodríguez, I.; Cela, R. 2010. Solid-phase microextraction with simultaneous

oxidative sample treatment for the sensitive determination of tetra- to hexa-

brominated diphenyl ethers in sediments. J. Chromatogr. A 1217, 14-21.




J. Sep. Sci. 26, 75-86.

Stapleton, H.M. 2006. Instrumental methods and challenges in quantifying polybrominated

diphenyl ethers in environmental extracts: A review. Anal. Bioanal. Chem. 386, 807-

817.

Talsness, C.E. 2008. Overview of toxicological aspects of polybrominated diphenyl ethers: A

flame-retardant additive in several consumer products. Environ. Res. 108, 158-167.

Vonderheide, A.P.; Mueller, K.E.; Meija, J; Welsh, G.L. 2008. Polybrominated diphenyl

ethers: Causes for concern and knowledge gaps regarding environmental

distribution, fate and toxicity. Sci. Total Environ. 400, 425-436.

Zhang, B.-Z.; Ni, H.-G.; Guan, Y.-F.; Zeng, E.Y. 2010. Occurrence, bioaccumulation and

potential sources of polybrominated diphenyl ethers in typical freshwater cultured

fish ponds of South China. Environ. Pollut. 158, 1876-1882.

CAPÍTULO 5

DESARROLLO DE METODOLOGÍAS SCREENING BASADAS EN CROMATOGRAFÍA DE GASES

ACOPLADA A ESPECTROMETRÍA DE MASAS CON ANALIZADOR DE TIEMPO DE VUELO PARA

LA IDENTIFICACIÓN DE CONTAMINANTES ORGÁNICOS.


131

5.1. Introducción


132


133

El mar, los ríos y, en general, todas las aguas superficiales han sido usadas

tradicionalmente como medio de evacuación de los residuos humanos. Los ciclos biológicos

del agua aseguran la reabsorción de estos restos orgánicos. Actualmente, la cantidad de

materia orgánica generada por el hombre sobrepasa, en ocasiones, el potencial depurador

de los mecanismos naturales. Además, ya no son solamente estos residuos orgánicos los

que son arrojados a los ríos y a los mares sino que también se desechan productos químicos

nocivos que destruyen la vida animal y vegetal acuática. Estos compuestos orgánicos

pueden llegar a depositarse en cualquier organismo o superficie asociado al agua, por lo

que es probable la presencia de diferentes familias de compuestos orgánicos, distribuidos

en el medio ambiente.

El presente capítulo muestra el gran potencial del acoplamiento GC-TOF MS para

investigar la presencia de contaminantes orgánicos en muestras ambientales. En los últimos

años, el uso de instrumentos de elevada resolución de masa como el TOF-MS ha ido

creciendo paulatinamente, siendo utilizado y reconocido como instrumento de medida en

el análisis de muestras ambientales, biológicas y alimentos. Esta técnica ofrece la

posibilidad de estudiar la presencia de compuestos target (compuestos diana) y también de

obtener información estructural sobre compuestos non-target (compuestos desconocidos)

presentes en la muestra analizada. Para ello, los analizadores TOF MS presentan una gran

sensibilidad y velocidad de adquisición en modo de adquisición de espectros completos (full

spectra acquisition) junto a medidas de masa con elevada exactitud (errores de masa del

orden de mDa).

Desde el punto de vista analítico, se pretende afrontar la investigación de

contaminantes orgánicos, trabajando con GC-TOF MS, mediante tres metodologías de

screening:

• Target screening: los analitos son seleccionados previamente a la adquisición por

MS. Únicamente se identifican los compuestos seleccionados a priori.


134

• Post-target screening: se adquiere el espectro completo de todos los compuestos

eluidos del sistema cromatográfico mediante MS. Los fragmentos iónicos

seleccionados se extraen a partir del cromatograma completo de la muestra.

• Non-target screening: El objetivo es investigar la presencia de compuestos de los

que se desconoce su identidad y presencia en las muestras problema.

En la primera parte de este capítulo, se aplica metodología “non-target” a partir de la

cual se pretende realizar una búsqueda de contaminantes orgánicos, no seleccionados

previamente, y que pueden estar presentes en las muestras de análisis (apartado 5.2).

En la segunda parte del capítulo se va a desarrollar y evaluar una metodología target

para el estudio de OPEs en sales y posteriormente la metodología desarrollada se aplica al

análisis de éstos en muestras de sal y muestras acuosas mediante GC-TOF MS (apartado

5.3).

Capítulo 5. Screening en muestras de sal y agua de mar

135

5.2. Screening de contaminantes orgánicos en muestras de sal y agua de mar


136


137


En este capítulo, se ha empleado el acoplamiento GC-TOF MS para la caracterización

de la contaminación presente en sales marinas procedentes de salinas de la costa

mediterránea. Tal y como se explicó en la introducción de esta Tesis, el detector TOF MS

proporciona una alta sensibilidad y elevada resolución de masa proporcionando espectros

de iones completos medidos con buena exactitud de masa (<3mDa). Las cualidades de

estos equipos sofisticados han permitido aplicaciones relevantes dentro de estudios

medioambientales tanto con GC o LC obteniendo resultados satisfactorios (Ferrer et al.,

2003; Čajka et al., 2004; Portolés et al., 2007). Recientemente los analizadores de masas

TOF han sido utilizados para el screening de plaguicidas en muestras ambientales (Feo et

al., 2010; Grimalt et al., 2010). Estos trabajos resaltan la capacidad de los analizadores TOF

para trabajar con diferentes enfoques analíticos (target, non-target). Uno de los

inconvenientes de este tipo de analizadores es su elevado precio y la necesidad de ser

manipulado por personal realmente especializado. También su menor rango lineal para la

cuantificación, así como su menor sensibilidad en comparación con los analizadores QqQ en

modo SRM, podrían ser inconvenientes a añadir. Estas deficiencias parecen mejorarse con

los analizadores TOF de nueva generación.

Hoy día, han sido publicadas metodologías analíticas suficientemente selectivas y

precisas para la identificación de compuestos diana (análisis target) mediante la utilización

de un analizador TOF (Dallüge et al., 2002; Hernández et al., 2009). Estas metodologías

permiten una identificación inequívoca de la presencia de estos compuestos en las

muestras estudiadas. Es por ello que se considera una herramienta muy adecuada para el

análisis de muestras procedentes del campo medioambiental, biológico y alimentario, que a

menudo contienen cientos, o aún miles, de compuestos orgánicos.

Actualmente, la lista de compuestos analizados y controlados mediante

metodologías target equivale a una fracción muy pequeña de la amplia lista de

contaminantes químicos existentes (Muir et al., 2006; Howard et al., 2010). Las

prestaciones de un analizador TOF permiten ampliar nuestro conocimiento sobre la posible


138

presencia de estos contaminantes en muestras ambientales, obteniendo resultados

satisfactorios. El procesamiento y tratamiento de datos con el analizador TOF es complejo

ya que se genera mucha información que debe ser interpretada correctamente.

Para el análisis de compuestos desconocidos, el analizador TOF presenta una serie de

herramientas de procesamiento de datos que permiten buscar, comparar e identificar

rápidamente los compuestos en matrices complejas. Este procesamiento de datos se basa

en una serie de algoritmos que permiten una exploración de datos exhaustiva para obtener

un nivel máximo de fiabilidad en la identificación. TOF MS permite la detección de un gran

número de compuestos y garantiza una elevada fiabilidad para la identificación de éstos

gracias a la adquisición de información espectral completa.

En el artículo que se incluye a continuación, se presenta una aplicación non-target

para el estudio de la posible presencia de contaminantes orgánicos en muestras salinas y de

agua de mar mediante la técnica GC-TOF MS.


139

5.2.2. Artículo de investigación 3.

Non-target screening of organic contaminants in marine salts by gas chromatography

coupled to high-resolution time-of-flight mass spectrometry

R. Serrano, J. Nácher-Mestre, T. Portolés, F. Amat, F. Hernández



140


141

Non-target screening of organic contaminants in marine salts by gas

chromatography coupled to high-resolution time-of-flight mass

spectrometry

Roque Serrano

a, Jaime Nácher-Mestre

a, Tania Portolés

a, Francisco Amat

b, Félix

Hernándeza*

aResearch Institute for Pesticides and Water (IUPA). Avda. Sos Baynat, s/n. University Jaume

I, 12071 Castellón, Spain bInstitute of Aquaculture of Torre la Sal (IATS), C.S.I.C., 12595 Ribera de Cabanes, Castellón,

Spain

Abstract

Gas chromatography coupled to high-resolution time-of-flight mass spectrometry (GC-TOF

MS) has been applied to characterize the organic pollution pattern of marine salt samples

collected in saltworks from the Spanish Mediterranean coast. After dissolving the samples

in water, the solution was subjected to solid-phase extraction reaching an overall 250-

preconcentration factor. The screening proposed allowed detecting sample components,

which identity was established by accurate mass measurements and by comparison of the

full-acquisition spectrum with theoretical MS libraries. Several organic pollutants were

identified in the marine salt samples. Some plasticizers, potentially toxic to humans, and

fragrances, included among the group of pharmaceuticals and personal care products were

detected. The contamination pattern present in the samples suggests a notable input of

pollutants from industries, farms and/or urbanized areas running off into the coastal

environment where sea water is pumped. The presence of these pollutants in the marine

salt samples suggests that they remain after the crystallization process. GC-TOF MS was

proven to be a powerful technique for wide-scope screening of organic contaminants in

order to monitoring the quality of marine salts thanks to the sensitive full-spectrum

acquisition at accurate mass, which facilitates the reliable identification of many GC-MS

ionizable compounds in the samples.

Keywords: marine salt, saltworks, saltpans, gas chromatography, mass spectrometry, non-

target screening.

* Corresponding author. Tel. +34-964-387366; e-mail address: [email protected]


142

1. Introduction

Marine salt is obtained by evaporation of sea water due to the combined effect of wind

blow and sunlight heat in the solar saltworks. Saltpans are located near the sea, becoming

peculiar environments inhabited by wildlife species associated to high salinity conditions.

Concern has arisen as consequence of the vulnerability of these environments to

anthropogenic pollution. Run-off from farms and industries may contain high

concentrations of pesticides and industrial sub-products and reach these vulnerable coastal

locations, with a deleterious impact on the briny aquatic systems (Ferrando et al., 1991).

This fact can also affect to the quality of the marine salt produced. Several authors have

reported the presence of contaminants in coastal sea and saline waters, such as pesticides

(Hernández et al., 1996), halocarbons, aliphatic and aromatic hydrocarbons and ketones

(Silva et al., 2009). Badoil and Benanou et al. (2009) have detected phenols, phosphates

and other volatile and semi-volatile compounds in waste landfill leachates, which reach

coastal waters. Contaminants produced by anthropogenic activities are transported by

rivers and water flows from wastewater treatment plants and are frequently deposited on

coastal locations like salt marshes or river estuaries and deltas. Several authors have

detected a variety of contaminants in these vulnerable areas (Wang et al., 2001; Navarro et

al., 2009). The marine salts obtained from saltpans can contain the contaminants present in

sea water, provided that they remain after the concentration and crystallisation processes.

As a consequence, monitoring the presence of organic contaminants in marine salts seems

necessary to have a realistic knowledge of their quality.

Hyphenation of gas chromatography (GC) with mass spectrometry (MS) is the most

widely used and accepted technique for determination of volatile and semivolatile

compounds of low-medium polarity in aquatic ecosystems, particularly in surface coastal

water and marine environments. Different MS analyzers have been applied for this purpose,

from single quadrupole to ion-trap or triple quadrupole, although the two later allow

working under tandem MS mode (Hernández et al., 1996; Silva et al., 2009; Bravo-Linares et

al., 2007; Pitarch et al., 2007). Recently, Silva et al. (2009) reported a methodology based


143

on head space solid phase microextraction and GC-quadrupole mass spectrometry for the

analysis of volatile compounds in marine salt, able to detect 40 volatile compounds

belonging to different chemical groups.

The wide majority of methods reported until now in the environmental field are

focused on a limited list of target contaminants. Even in the case that target pollutants

investigated belong to priority lists, target methods do not allow the wide-scope screening

required to investigate a large number of compounds that might be present in the samples.

In most target methods, other non-selected contaminants would not be detected due to

the specific-analyte information acquired. Although conventional MS analyzers can also

work under scan mode, their capability to detect organic contaminants at low levels in

complex-matrix samples is rather limited due to their low sensitivity and selectivity and

their nominal mass measurements.

The recent irruption of modern high-resolution time-of-flight (TOF) analyzers opens

new perspectives to develop wide-scope screening methodologies. GC-TOF MS offers

interesting features for this purpose, as it combines high full-spectrum sensitivity and

elevated mass resolution making feasible the accurate mass measurements of the

molecular and/or fragments ions of any GC-amenable compound present in the sample.

This technique allows searching organic contaminants in a post-target (i.e. searching for

selected compounds after MS acquisition) and also in a non-target way (i.e. searching for

unknowns, without any kind of compounds selection) (Hernández et al., 2010). GC-TOF MS

has been successfully applied for screening, identification and elucidation of organic

pollutants in environmental water and biological samples (Hernández et al., 2007, 2009),

and also for confirmation of pollutants in highly complex matrix like wastewater (Ellis et al.,

2007).

In this work, we have applied GC-TOF MS for the rapid and wide-scope screening of

organic pollutants in sea water and in marine salts obtained from solar saltworks and from

a pristine sea shore salt marsh sited along the Spanish Western Mediterranean coast. The

identity of the sample components detected in a non-target way was established by means

of exact mass measurements and by comparison with theoretical spectral libraries. In


144

addition, the organophosphate esters (OPEs) identified were confirmed by injecting

reference standards.

2. Material and methods

2.1. Sampling points.

Marine salt samples from four solar saltworks sited in the Spanish Mediterranean shore

(see Figure 1) were collected directly from the crystallized salt stock in saltpans (samples 3

and 5) or purchased from the producers (samples 1 and 4). A seawater sample was also

collected from the sea shore in front of a pristine salt marsh located in Torre la Sal,

neighbouring a natural protected area (Natural Park of Ribera de Cabanes, Spain), sited

close to the city of Castellon (Sampling point 2). Sampling point 1 is a solar saltwork sited in

the Alfaques bay, south of the Ebro River delta. This river receives domestic and industrial

wastewater from numerous minor settlements along its way. Discharges into the Ebro River

vary at different locations, showing an increase downstream, probably due to inputs from

the tributaries or natural recharge of the stream, and finally it flows into the Mediterranean

sea after crossing through the Ebro Delta (Bouza-Deaño et al., 2008). Sampling point 3 is a

solar saltwork located in the vicinity of an important fishing and middle trade harbour,

surrounded by a highly urbanized area. Sampling points 4 and 5 are solar saltworks sited in

high valuable natural areas but neighbouring important summer touristic areas. All samples

were stored at -20°C until analysis.

2.2. Reagents.

HPLC-grade water was obtained from a MilliQ water purification system (Millipore Ltd.,

Bedford, MA, USA). Acetone, Ethyl Acetate, Dichlorometane (DCM) and n-Hexane (ultra

trace quality) used in solid-phase extraction (SPE) experiments were purchased from

Scharlab (Barcelona, Spain). Bond Elut cartridges C18 (500 mg) (Varian, Harbor City, CA,

USA) were used for SPE. Triphenyl phosphate (TPhP) and 2-Ethylhexyl diphenyl phosphate

(EHDPP) reference standards were purchased from TCI Europe (Zwijndrecht, Belgium). Tri-


145

n-buthyl phosphate (TBP) and Tris(1-chloro-2-propyl) phosphate (TCPP) reference

standards were purchased from Sigma-Aldrich (Madrid, Spain).

2.3. GC-TOF MS instrumentation.

GC system (Agilent 6890N; Agilent Palo Alto, USA) equipped with an autosampler (Agilent

7683) was coupled to a time-of-flight mass spectrometer (GCT, Waters Corporation,

Manchester, U.K.), operating in electron ionization (EI). GC separation was performed using

a fused silica HP-5MS capillary column with a length of 30 m, an internal diameter of 0.25

mm and a film thickness of 0.25 μm (J&W Scientific, Folson, CA, USA). The injector

temperature was set to 280°C. Splitless injections of 1 μL samples were carried out. Helium

(99.999%; Carburos Metálicos, Valencia, Spain) was used as carrier gas at a flow rate of 1

mL/min. The interface and source temperature were set to 250°C and a solvent delay of 4

min was selected.

The oven program in GC-TOF MS analysis was programmed as follows: 90 °C (1min);

5 °C/ min to 300 °C (2 min). The TOF MS operated at 1 spectrum/s, acquisition rate over the

mass range m/z 50-650, using a multichannel plate voltage of 2850 V. TOF-MS resolution

was approximately 7000 (FWHM). Heptacosa standard, used for the daily mass calibration

and as lock mass, was injected via syringe in the reference reservoir at 30°C for this

purpose; the m/z ion monitored was 218.9856. The application manager ChromaLynx and

TargetLynx was used to process the qualitative data obtained from standards and from

sample analysis. Library search was performed using the commercial NIST library.

2.4. Recommended analytical procedure.

The sample procedure was based on the previously applied for the determination of around

50 compounds, including organochlorine and organophosphorus insecticides, herbicides,

polychlorinated biphenyls, polyciclic aromatic hydrocarbons, brominated diphenyl ethers,

octyl/nonyl phenols and pentachlorobenzene with some modifications (Pitarch et al.,

2007). Briefly, 62.5 g of salt were diluted with water to a final volume of 250 mL and


146

filtered. The filtered solution was passed through the C18 SPE cartridge, previously

conditioned by passing 6 mL methanol, 6 mL ethyl acetate:DCM (50:50), 6 mL methanol and

6 mL water avoiding dryness. After loading the sample (250 mL), the cartridges were

washed with 3 mL water and dried by passing air under vacuum for at least 15 min. The

elution was performed by passing 5 mL ethyl acetate:DCM (50:50). The extract collected

was evaporated under a gentle nitrogen stream at 40°C and redisolved in 0.25 mL of n-

hexane. The overall procedure also involved a method blank to test that no contamination

was introduced in the extracts along the analysis.

2.5. GC-TOF MS methodology for non-target screening.

GC-TOF MS non-target screening was carried out by using the ChromaLynx Application

Manager. This software was used to detect the presence of multiple components and to

show its deconvoluted MS spectra to be submitted to library search routine (in our case

NIST library). Components are reduced to a list of possible candidates by using the list

factor from the mass library search (library match >700). Then, accurate mass confirmation

is automatically performed. The formula from the library list is submitted to an Elemental

Composition calculator and accurate mass measurements of (up to) 5 abundant ions are

evaluated for confirmation/rejection of the finding (for more details see Hernández et al.,

2007, 2010).

3. Results and Discussion

The analytical methodology described was applied to the analysis of one sea water and four

marine salt samples collected from different solar saltworks located along the Spanish

Mediterranean coast. The sensitive and reliable qualitative analysis was feasible thanks to

the 250-fold pre-concentration in the sample, with low sample handling as corresponds to

the SPE procedures, which was combined with the advantages offered by GC-TOF MS. The

procedure applied allowed the confirmation of the compounds’ identity of at estimated

concentrations around sub-ng/Kg, which was tested for some reference standards of the


147

compounds detected. As shown in this paper and confirmed in our previous works

(Hernández et al., 2007, 2009), the non-target methodology applied for screening organic

contaminants is able to detect and identify a large number of GC-amenable compounds

belonging to different chemical families. However, a genuine non-target analysis is a

laborious and time-consuming task, as a consequence of the huge amount of

chromatographic peaks from the sample components and to the lack of list of compounds

to be searched. Therefore, the use of advanced processing software is required to facilitate

this task. This software should be able to detect relevant/abundant sample components

and to confirm their identity making use of the accurate full-spectrum data provided by TOF

MS. Although a part of the process can be performed (almost) in an automated way, the

expertise of the analyst on MS spectra interpretation and the knowledge of the MS

fragmentation rules are needed for a successful analysis (Hernández et al., 2010).

The samples analyzed contained volatile and semi volatile compounds, including

industrial sub-products, pesticides, flame retardants, plasticizers and personal care

products (Table 1). This kind of contaminants have been also found in other studies related

to water pollution, and they are into the environment as a consequence of anthropogenic

activities (Badoil et al., 2009; Navarro et al., 2009; Hernández et al., 2007; Martínez-

Carballo et al., 2007; Reemtsma et al., 2008; Matamoros et al., 2009).

Table 1 shows the contamination pattern observed in the marine salt and seawater

samples studied in this work. The seawater sample collected from the sea shore at Torre la

Sal (S2), considered as a protected natural area relatively free of contaminant sources, was

almost free of the contaminants found in marine salt, and only two alkyl phenols and one

organic acid were identified. On the contrary, the marine salt samples were more

contaminated. The type of compounds detected seemed to vary according to the

geographical location of the saltwork.

As an illustrative example, Table 2 shows the confirmation of identity of the

compounds detected in “Santa Pola” salt sample (S3). The elemental composition could be

proposed for at least 4 m/z fragment ions based on accurate mass measurements. In

addition, the experimental accurate mass for the main ions was compared with the


148

theoretical ones. In general, mass errors were below 3 mDa, except for a few low-abundant

ions. An example of the non-target detection of TCPP in “San Pedro del Pinatar” salt sample

is given in Figure 2. Five ions were selected from the EI spectrum for the accurate mass

confirmation of the identity of TCPP, with mass errors always below 2.6 mDa. In addition,

the chemical structure suggested for these ions was in agreement with that of TCPP.

It is worth to notice that several of the compounds detected belong to the OPEs

family. These chemicals are produced in large quantities for their use as flame retardants,

plasticizers and also as pesticides. Their widespread use and presence in host materials led

to a continuous discharge and distribution through wastewaters (Reemtsma et al., 2008),

and coastal areas are the fate of wastewaters from industrial and urban activities

containing these and other pollutants. As a consequence of the toxicity and environmental

persistence of OPEs, their presence in marine salt intended for human consumption should

be under control to assure its right quality.

Considering the interest of OPEs, reference standards of TCPP, TBP, TPhP and EHDPP

were acquired in a subsequent step to perform additional experiments for confirmation.

We could not find the reference standard of bis(1-chloro-2-propyl) (3-chloro-1-

propyl)phosphate, which was also detected in the non-target screening. Using reference

standards it was feasible to test the retention time and to obtain their TOF MS spectrum to

unequivocally confirm the presence of these compounds in the samples. The experiments

with reference standards allowed us to confirm all positives previously reported by TOF MS,

demonstrating the excellent potential of this technique for identificative purposes, even

without reference standards.

As an illustrative example, Figure 3 shows the eXtracted Ion Chromatograms (XICs)

for the positive of EHDPP detected in “La Trinitat Saltwork” salt sample (S1) which could be

additionally confirmed using the reference standard. The presence of the chromatographic

peaks in the XICs, at the expected retention time, and the attainment of all Q/q ratios when

comparing with the reference standard allowed the confirmation of the finding in the

sample. The corresponding EI accurate mass spectra generated by TOF MS are also shown.

Mass errors for four representative ions were below 3.2 mDa, which gave more confidence


149

to the confirmation process. Chemical structures for the most abundant fragment ions

were suggested based on the elemental compositions proposed accordingly to the accurate

mass measurements given by the instrument.

Apart from OPEs, the most abundant compounds detected were alkyl phenols.

Fragrances and plasticizers were also identified in some salt samples. The presence of

alkylphenols in aquatic environments has been previously reported by several authors

(Hernández et al., 2010; Sheldon et al., 1978; During et al., 2002). They are degradation

products from alkylphenol polyethoxylates, mainly applied to pesticide formulations and as

plastic additives, among other uses (Badoil et al., 2009). The persistence and accumulation

properties of alkylphenols have led to their wide distribution in different environmental

compartments (Vazquez-Duhalt et al., 2006; Correa-Reyes et al., 2007). The sources of

these pollutants are commonly the wastewaters from industrial and municipal treatment

plants (USEPA, 2005) and their accumulation has been observed in sediments receiving

contaminated water flows (Petrovic et al., 2002). Data obtained in the present work suggest

that these compounds have been accumulated in salts after the crystallization process. The

presence of these pollutants might pose a threat to the quality of the salt produced in

saltworks sited in environments like deltaic and estuarine locations receiving water flows

from industrial and/or urbanized areas. In fact, most detections of alkylphenols

corresponded to sampling points 1 and 3 (which accomplish these characteristics; see

description in Experimental section). Recently, Navarro et al. (Navarro et al., 2009), making

use of GC-MS with single quadrupole, have detected several of these compounds in the

Ebro River sediments, in which delta the sampling point 1 is sited, as indicated above.

The presence of Di-(2ethylhexyl)adipate in marine salts is also of concern. This

compound is used as plasticizer for food packaging, and it is considered as carcinogen and

endocrine disruptor (Rahman et al., 2009; Ghisari et al., 2009). Another plasticizer detected,

and also considered as endocrine disruptor, was benzyl butyl phthalate. This compound has

been previously reported to be present in marine sediments (Xu et al., 2009). Butylated

hydroxytoluene (BHT) was detected also in samples 3 and 4. This compound is an

antioxidant widely used as food aditive and in biological samples for storage before


150

analysis, as well as in cosmetics, pharmaceuticals, jet fuels, among other uses (Badoil et al.,

2009), and it has been found in aquatic environments (Hernández et al., 2007; Fries et al.,

2002; Gómez et al., 2009).

2-Oxohexamethylenimine (caprolactam) -the monomer of nylon-6- has been

identified in marine salt samples probably due to the use of ammonium sulphate (a sub

product obtained during the manufacture of the polymer) in growing crops as fertilizer.

Methyl dihydrojasmonate and galaxolide were other compounds detected. They are used

as fragrances, and are included in the group of water contaminants called Pharmaceuticals

and Personal Care products (PPCPs), which are suspected to be an environmental problem

still not well known (Daughton et al., 1999). Similarly to other organic contaminants, these

compounds could be removed entirely or partly by means of adequate technologies of

wastewater treatment (Matamoros et al., 2009). Dihydroactinidiolide, detected in sample

4, is a volatile terpene occurring naturally in a variety of plants and insects, but it has also

been prepared synthetically for its use as a fragrance (Yao et al., 1998).

Other relevant compounds detected in marine salt were benzophenone and 3-

methyl-benzophenone, used as photoinitiator in UV-curing applications and as UV filter

(Badoil et al., 2009); cyclic octaatomic sulfur, indicator of microbiological activity (Badoil et

al., 2009); and nonanoic acid, used in the preparation of plasticizers and lacquers, and also

as herbicide.

All identifications reported in this work were supported by accurate mass

measurements of several EI ions (up to five in most of the cases), by the low mass errors

observed in relation to their theoretical exact masses, and by the compatibility of the

chemical structures proposed for these ions with the chemical structure of the compound

identified.

The contamination pattern observed in the marine salt samples includes up to 25

organic compounds, with around 12 of them being present in every sample. Sources of

these contaminants surely are run offs from industries, farms and urbanized areas. Our

findings suggest an important input of these pollutants into the environments around the


151

saltpans, which is in agreements with data reported in similar areas (Daughton et al 1999;

Ternes et al., 2004; Muñoz et al., 2008). The presence of the compounds identified in the

marine salt samples indicates that they are concentrated and that they persist along the

crystallization process.

4. Conclusion

Without using any previous list of compounds to be investigated, the non-target

methodology applied in this work has allowed the detection and reliable identification of

several relevant contaminants of anthropogenic origin, belonging to quite different

chemical groups. The strong potential of GC-TOF MS for qualitative purposes comes from

the full spectrum acquisition at accurate mass, with satisfactory sensitivity, provided by this

instrument. Making an appropriate use of all relevant information given by this technique it

has been feasible to identify many contaminants in a reliable way, even without reference

standards being available, as illustrated in this work. Surely, several of the compounds

detected in marine salt would not had been detected using a target approach, as although

relevant they might not have been included in a target screening, which is typically focused

on a limited list of priority pollutants.

In the light of the results reported, we can conclude that priority pollutants, typically

subjected to strict control, constitute only part of the large chemical pollution puzzle. There

is a diverse group of unregulated pollutants, including industrial sub-products, PPCPs, and

an increasing concern on the risks that they pose on humans and on the environment.

Acknowledgments

This research was supported by the Spanish Ministry for Education and Science (Projects

CGL2005-02306/BOS and CTQ 2009-12347). The authors acknowledge the financial support

of the Generalitat Valenciana, as research group of excellence PROMETEO/2009/054. The

authors are also grateful to the Serveis Centrals d’Instrumentació Científica (SCIC) of the

University Jaume I for the use the GC-TOF MS instrument.


152

References

Badoil, L., Benanou, D., 2009. Characterization of volatile and semivolatile compounds in

waste landfill leachates using stir bar sorptive extraction-GC/MS. Analytical and

Bioanalytical Chemistry 393, 1043-1054.

Bravo-Linares, C. M., Mudge, S. M., Loyola-Sepulveda, R. H., 2007. Ocurrence of volatile

organic compounds (VOCs) in Liverpool Bay, Irish Sea. Marine Pollution Bulletin 54,

1742-1753.

Bouza-Deaño R., Ternero-Rodríguez, M., Fernández-Espinosa A. J., 2008.Trend study and

assessment of surface water quality in the Ebro River (Spain). Journal of Hydrology

361, 227-239.

Correa-Reyes, G., Viana, M. T., Marquez-Rocha, F. J., Licea, A. F., Ponce, E., Vazquez-Duhalt,

R., 2007. Nonylphenol algal bioaccumulation and its effect through the trophic chain.

Chemosphere 68, 662-670.

Daughton C. G., Ternes T. A., 1999. Pharmaceuticals and personal care products in the

environment: agents of subtle change. Environmental Health Perspectives 107, 907-

938.

During, R. A., Krahe, S., Gath, S., 2002. Sorption Behaviour of nonylphenol in terrestrial

soils. Environmental Science & Technology 36, 4052-4057.

Ellis, J., Shah, M., Kubachka, K. M., Caruso, J. A., 2007. Determination of organophosphorus

fire retardants and plasticizers in wastewater samples using MAE-SPME with GC-

IPCMS and GC- TOF MS detection. Journal of Environmental Monitoring 9, 1329-

1336.

Ferrando, M. D., Andreu-Moliner, E., 1991. Acute lethal toxicity of some pesticides to

Brachionus calyciflorus and Brachionus plicatilis. Bulletin of Environmental

contamination and Toxicology 47, 479-484.

Fries, E., Püttmann, W., 2002.Analysis of the antioxidant butylated hydroxytoluene (BHT) in

water by means of solid phase extraction combined with GC/MS. Water Research 36,

2319-2327.


153

Ghisari, M., Jorgensen, E. C. B., 2009.Effects of plasticizers and their mixtures on estrogen

receptor and thyroid hormone functions. Toxicology Letters 189, 67-77.

Gómez, M. J., Gómez-Ramos, M. M., Agüera, A., Mezcua, M., Herrera, S., Fernández-Alba,

A.R., 2009. A new gas chromatography/mass spectrometry method for the

simultaneous analysis of target and non-target organic contaminants in waters.

Journal of Chromatography A 1216, 4071-4082.

Hernández, F., Serrano, R., Miralles, M. C., Font, N., 1996.Gas and liquid chromatography

and enzyme linked immunosorbent assay in pesticide monitoring of surface water

from the Western Mediterranean (Comunidad Valenciana, Spain). Chromatographia

42, 151-158.

Hernández, F., Portolés, T., Pitarch, E., Lopez, F. J., 2007.Target and non target screening of

organoc micropollutants in water by solid-phase microextraction combined with gas

chromatography/high-resolution time-of-flight mass spectrometry. Analytical

Chemistry 79, 9494-9504

Hernández, F., Portolés, T., Pitarch, E., López, F. J., 2009. Searching for anthropogenic


flight mass spectrometry. Journal of Mass Spectrometry 44, 1-11.

Hernández, F., Portolés, T., Pitarch, E., Lopez, F. J., 2010. Gas Chromatography/High-

resolution mass spectrometry: An advanced analytical tool to investigate the

presence of organic compounds at trace levels in environmental, food safety and

toxicology fields. Submitted to TRAC-Trends in Analytical Chemistry. September.

Martínez-Carballo, E., González-Barreiro, C., Sitka, A., Scharf, S., Gans, O., 2007.

Determination of selected organophosphate esters in the aquatic environment of

Austria. Science of the Total Environment 388, 290-299.

Matamoros, V., Arias, C., Brix, H., Bayona, J. M., 2009. Preliminary screening of small-scale

domestic wastewater treatment systems for removal of pharmaceuticals and

personal care products. Water Research 43, 55-62.

Muñoz, I., Gómez, M. J., Molina-Díaz, A., Huijbregts, M. A. J., Fernández-Alba, A. R., García-

Calvo, E., 2008. Ranking potential impacts of priority and emerging pollutants in

urban wastewater through life cycle impact assessment. Chemosphere 74, 37-44.


154

Navarro, A., Endo, S., Gocht, T., Barth, J. A. C., Lacorte, S., Barceló, D., Grathwohl, P., 2009.

Sorption of alkylphenols on Ebro River sediments: Comparing isotherms with field

observations in river water and sediments. Environmental Pollution 157, 698-703.

Petrovic, M., Lacorte, S., Viana, P., Barceló, D., 2002. Pressured liquid extraction followed by

liquid chromatography-mass spectrometry for the determination of alkylphenolic

compounds in river sediment. Journal of Chromatography A 959, 15-23.

Pitarch, E., Medina, C., Portolés, T., López, F.J., Hernández, F., 2007. Determination of

priority micro-polutants in water by gas chromatography coupled to triple

quadrupole mass spectrometry. Analytica Chimica Acta 583, 246-258.

Rahman, M. B. A., Chaibakhsh, N., Basri, M., Salleh, A. B., Rahman, R. N. Z. R. A., 2009.

Application of artificial network for yield prediction of lipase-catalyzed synthesis of

dioctyl adipate. Applied Biochemistry and Biotechnology 158, 722-735.

Reemtsma, T., Quintana, J. B., Rodil, R., Garcia-Lopez, M., Rodriguez, I., 2008.

Organophosphorus flame retardants and plasticizers in water and air. I. Occurrence

and fate. TRAC-Trends in Analytical Chemistry 27, 727-737.

Sheldon, L. S., Hites, R. A., 1978. Organic compounds in the Delaware River. Environmental

Science & Technology 12, 1188-1194.

Silva, I., Rocha, S. M., Coimbra, M. A., 2009. Head solid phase microextraction and gas

chromatography-quadrupole mass spectrometry methodology for analysis of volatile

compounds of marine salt as potential origin biomarkers. Analytica Chimica Acta 635,

167-174.

Ternes, T. A., Joss A., Siegrist, H., 2004.Scrutinizing pharmaceuticals and personal care

products in wastewater treatment. Environmental Science & Technology 38, 392-

399.

USEPA. 2005. Aquatic Life Ambient Water Quality Criteria – Nonylphenol. U.S.

Environmental Protection Agency, Office of Water: Washington, DC.

Vazquez-Duhalt, R., Marquez-Rocha, F. J., Ponce, E., Licea, A. F., Viana, M. T., 2006.

Nonylphenol, an integrated vision of a pollutant. Applied Ecology and Environmental

Research 4, 1-25.


155

Wang, Z., Fingas, M., Owens, E. H., Sigouin, L., Brown, C. E., 2001. Long-term fate and

persistence of the spilled Metula oil in a marine salt marsh environment. Degradation

of petroleum biomarkers. Journal of Chromatography A. 926, 275-290.

Xu, X-R., Li, X. Y., 2009. Sorption behaviour of benzyl butyl phthalate on marine sediments:

Equilibrium assessments, effects of organic carbon content, temperature and salinity.

Marine Chemistry 115, 66-71.

Yao, S., Johannsen, M., Hazell, R. G., Jorgensen, K. A., 1998. Total synthesis of (R)-

Dihydroactiniolide and (R)-Actiniolide using asymmetric catalytic hetero-diels-alder

methodology. Journal of Organic Chemistry 63, 118-121.


156

Table 1. Compounds identified in marine salt and seawater sample.

Compound CAS Number S1 S2(w) S3 S4 S5 Observationsa

1-[4-(1-methylethenyl)phenyl]-Ethanone 5359-04-6 X X X Industrial sub-product

2,4-di-tert-butylphenol 96-76-4 X X Toxic and dangerous for the environment, highly flammable, harmful

and irritant.

2-[(Z)-3-hydroxy-3-methyl-1-butenyl]phenol 17235-14-2 X Industrial sub-product

2-phenoxyethanol 122-99-6 X Anesthesic

2-oxohexamethylenimine (Caprolactam) 105-60-2 X X X X Toxic by ingestion, inhalation, or absorption through the skin.

3,5-di-tert-butylphenol 1138-52-9 X Antioxidants and light-protection agents

3,6-di-tert-butyl-4-ethylphenol 4130-42-1 X X X Non toxic

3-methyl-benzophenone 134-84-9 X X UV filter. UV-curing applications

4,5,7-trichloro-2-methyl- benzofuran 18628-11-0 X X Pesticide

4-tert-amylphenol 80-46-6 X Intermediate for organic mercury germicides pesticides and chemicals

used in rubber and petroleum industries

4 -tert-octylphenol 140-66-9 X X Acutely very toxic to aquatic organisms and may cause long-term

adverse effects in the aquatic environment

Benzophenone 119-61-9 X X UV filter. UV-curing applications

Benzyl butyl phthalate 85-68-7 X Plasticizer. Toxic effects such as cellular necrosis

Bis (1-chloro-2-propyl) (3-chloro-1-propyl)

phosphate

137909-40-1 X X Pesticide, toxic and irritating

BHT 128-37-0 X X Synthethic antioxidant

Cyclic octaatomic sulfur 10544-50-0 X Microbiological activity indicatora

Di (2-ethylhexyl) adipate 103-23-1 X human carcinogen

Dihydroactinidiolide 17092-92-1 X Volatile terpene (large structure hydrocarbon)

EHDPP 1241-94-7 X X Pesticide, toxic to aquatic organisms

Galaxolide 1222-05-5 X X Musk fragrance

Methyl dihydrojasmonate 24851-98-7 X X X Musk fragrance

Nonanoic acid 112-05-0 X Irritant

TBP 126-73-8 X X X X Used as a herbicide and fungicide

TCPP 13674-84-5 X X X X Pesticide, flame retardant

TPhP 115-86-6 X X Pesticide, plasticizer and flame retardant

S1: Delta del Ebro (La Trinitat Saltworks), Tarragona; S2(w): Sea water from Torre la Sal Sea shore, Castellón; S3: Santa Pola Saltworks, Alicante;

S4: Torrevieja Saltworks, Alicante; S5: San Pedro del Pinatar Saltworks, Murcia; a

International Chemical Safety Cards: www.inchem.org/documents


157

Table 2. Confirmation of organic compounds in S3 salt sample. Molecular peak Ion 1 Ion 2 Ion 3 Ion 4 Ion 5

Compound N° CAS

Molecular

formula

Molecular

mass

Elemental

composition

Experimental

m/z

(error in mDa)

Elemental

composition

Experimental

m/z

(error in mDa)

Elemental

composition

Experimental

m/z

(error in mDa)

Elemental

composition

Experimental

m/z

(error in mDa)

Elemental

composition

Experimental

m/z

(error in mDa)

2,4-di-tert-butylphenol 96-76-4 C14H22O 206.1671 C14H22O 206.1660

(-1.1)

C13H20O 192.1483

(-3.1)

C13H19O 191.1443 (0.7) C11H15O 163.1120(-0.3) C9H11O 135.0802

(-0.8)

3,6-di-tert-butyl-4-

ethylphenol 4130-42-1 C16H26O 234.1984 C16H26O 234.2013 (2.9) C15H24O 220.1836 (0.9) C15H23O 219.1760 (1.1) C6H6 78.0442 (-2.8)

4-tert-octylphenol 140-66-9 C14H22O 206.1671 C9H12O 136.0857

(-3.1)

C9H11O 135.0798

(-1.2)

C8H7O 119.0475

(-2.2)

C7H7O 107.0496(-0.1)

4-tert-amylphenol 80-46-6 C11H16O 164.1201 C10H13O 149.0969 (0.3) C9H11O 135.0810 (0) C8H9O 121.0666 (1.3) C7H7O 107.0486(-1.1) C6H7O 95.0477

(-2.0)

TCPP 13674-84-5 C9H18Cl3O4P 326.0008 C5H11O4PCl 201.0100 (1.6) C3H7O3PCl 156.9818

(-0.3)

C2H6O4P 125.0001

(-0.3)

H4O4P 98.9826 (-2.1) C2H5OP 76.0058

(-3.0)

TPhP 115-86-6 C18H15O4P 326.0708 C18H15O4P 326.0737 (2.9) C18H14O4P 325.0656 (2.6) C6H6O 94.0458 (3.9) C6H5 77.0375 (-1.6)

TBP 126-73-8 C12H27O4P 266.1647 C4H12O4P 155.0480 (0.7) C2H6O4P 125.0019 (1.5) H4O4P 98.9830 (-1.7) C4H8 56.0567 (-5.9) C8H20O4P 211.1109

(1.0 )

Bis (1-chloro-2-propyl)

(3-chloro-1-propyl)

phosphate

137909-40-1 C9H18Cl3O4P 326.0008 C3H7ClO3P 156.9818

(-0.3) C2H6O4P

125.0003

(-0.1) C2H3Cl2O3P 116.9509 (0.1) H4O4P 98.9828 (-1.9) C2H5OP

76.0048

(3.0)

EHDPP 1241-94-7 C20H27O4P 362.1647 C12H12O4P 251.0457(-1.6) C12H10O 170.0750 (1.8) C8H16 112.1253 (0.1) C6H6O 94.0412 (-0.7) C6H5 77.0366

(-2.5)

BHT 128-37-0 C15H24O 220.1827 C15H24O 220.1861 (3.4) C14H21O 205.1611 (1.9) C11H13O 161.0978 (1.2) C11H13 145.0983(-3.4)

2-oxohexamethylenimine

(Caprolactam) 105-60-2 C6H11NO 113.0841 C6H11NO 113.0822

(-1.9) C4H7NO 85.0555 (2.7) C5H8O 84.0547 (-2.8) C2H2NO 56.0169 (3.3) C2HNO 55.0116

(5.8)

Methyl

dihydrojasmonate

24851-98-7 C13H22O3 226.1569 C13H22O3 226.1601 (3.2) C8H12O3 156.0773(-1.3) C10H17O 153.1274(-0.5) C5H7O 83.0504 (0.7)

Nonanoic acid 112-05-0 C9H18O2 158.1307 C7H13O2 129.0920 (0.4) C6H11O2 115.0756(-0.3) C3H5O2 73.0258 (-3.2) C2H4O2 60.0182 (-2.9)


158

Figure captions

Figure 1. Area of study and sampling points

Figure 2. Detection of TCPP in the salt sample S5 by GC-TOF MS non-target screening. (A)

Extracted ion Chromatograms for five fragment ions. (B) Library mass spectrum of TCPP at

nominal mass. (C) Experimental EI accurate mass spectrum of the positive finding of TCPP.

Chemical structures proposed for the five most abundant EI fragment ions and mass errors.

Figure 3. GC-TOF MS extracted ion chromatograms (top) at different m/z (mass window

0.02 Da) and accurate mass spectrum (bottom) for EHDPP for the reference standard (left)

and for one positive salt sample (right). Q, qualitative ion; q, confirmative ion; St, reference

standard; S, sample; Q/q ratio within tolerance limits. Chemical structures proposed for the

most abundant fragment ions.


159

Valencia

Barcelona

Sampling point 1. La Trinitat Saltworks

Spain

Castellón

Valencia

Barcelona

Sampling point 5. San Pedro del Pinatar Saltwork

Sampling point 3. Santa Pola SaltworkSampling point 4. Torrevieja Saltwork

Sampling point 2. Torre la Sal

Spain

Castellón

Figure 1.


160

m/z100 150 200 250 300 350

%

0

100

3.818e+003Suggested compound: 2-Propanol, 1-chloro-, phosphate (3:1)

125.000198.9826

201.0100

277.0189203.0056

215.0230 281.0250

m/z

%

0

100

Library: 2-Propanol, 1-chloro-, phosphate (3:1); C9H18Cl3O4P

125

99

77

76

117157

139

201

175

177

277

203

215233

251

279

291295 315

min20.100 20.200 20.300 20.400

%

0

100

5.671e+0032-Propanol, 1-chloro-, phosphate (3:1)

A

B

C

TCPP

C9H18Cl3O4P

O

PO

OO CH3

CH3

CH3

Cl

Cl

Cl

TCPP

O

PO

OO

CH+

CH3

CH3

Cl

Cl

CH3

C8H16Cl2O4P2.6 mDa

O

POH

OO

CH+

CH3

CH3

Cl

C5H11ClO4P1.6 mDa

C3H7ClO3P1.1 mDa

O

POO

+

CH3

Cl

HC2H6O4P-0.3 mDa

O

P

OCH+

CH3

OH

OH

H4O4P-2.1 mDa

O

P

OH

OH

OH2+

156.9832

Suggested compound: TCPP

Library: TCPP

Library match= 817

Figure 2


161

Salt sample from La Trinitat Saltworks

Time30.00 32.00 34.00

%

0

100

30.00 32.00 34.00

%

0

100

30.00 32.00 34.00

%

0

100

30.00 32.00 34.00

%

0

100

MP246 TOF MS EI+ 251.047 0.02Da

1.02e3Area

31.58;253

MP246 TOF MS EI+ 250.039 0.02Da

304Area

31.58;75

MP246 TOF MS EI+ 249.031 0.02Da

235Area

31.58;58

MP246 TOF MS EI+ 170.073 0.02Da

136Area

31.58;34

EHDPP Standard 1 mg/L

Time30.00 32.00 34.00

%

0

100

30.00 32.00 34.00

%

0

100

30.00 32.00 34.00

%

0

100

30.00 32.00 34.00

%

0

100

MP257 TOF MS EI+ 251.047 0.02Da

250Area

31.63;64

MP257 TOF MS EI+ 250.039 0.02Da

82.2Area

31.63;21

MP257 TOF MS EI+ 249.031 0.02Da

60.4Area

31.63;15

MP257 TOF MS EI+ 170.073 0.02Da

41.1Area

31.63;10

Ion 1 (Q)

Ion 2 (q) Q/q ratio (St)= 3.04

Q/q ratio (S)= 3.37

11% �

Q/q ratio (St)= 4.26

Q/q ratio (S)= 4.36

2% �

Q/q ratio (St)= 6.40

Q/q ratio (S)= 7.44

16% �

Ion 3 (q)

Ion 4 (q)

Ion 1 (Q)

Ion 2 (q)

Ion 3 (q)

Ion 4 (q)

Octizicer 1ppm

m/z100 150 200 250 300 350 400

%

0

100

MP257 1718 (31.635) Cm (1716:1721-1741:1800) TOF MS EI+ 3.46e3251.0462

112.1267

94.0426

250.0400

249.0300

170.0772

112.1789

140.0033 232.0283

252.0506

264.9822362.1635

341.0565374.0013

S6

m/z100 150 200 250 300 350 400

%

0

100

MP246 1715 (31.586) Cm (1714:1716-(1640:1676+1751:1783)) TOF MS EI+ 1.27e4251.0491

250.0424

249.0336

94.0411

170.0764

169.0710

168.0620232.0308

251.1254

252.1359362.1724305.0947

402.0457

P

OOH

2

+

OO

C12H12O4P1.8 mDa

C12H11O4P2.9 mDa

P

OO

OH+O

P

O+

O

OO

C12H10O4P1.9 mDa

C6H6O-0.8 mDa

O

C+

CH

CH2

Salt sample from La Trinitat SaltworksEHDPP Standard 1 mg/L

P

OO

OO

CH3

CH3

O+

C12H10O3.2 mDa

Figure 3


163


La metodología analítica desarrollada fue llevada a cabo para el análisis de

contaminantes en muestras con elevado contenido salino, concretamente en sales marinas

y agua de mar. El tratamiento de las muestras se basó en una purificación y posterior

concentración mediante SPE. Partiendo de una muestra con elevado contenido salino

(aproximadamente 25% sal) se obtenía un extracto final concentrado 250 veces (ver Figura

5.1). Posteriormente, el extracto se introducía en el sistema GC-TOF MS para el screening

de contaminantes orgánicos.

GC-TOF MS

62.5 g muestra25%sal

+ Vf=250ml

SPEC18 (500 mg)

+ filtración 0.45µm

Extracto purificado5ml

+ corriente de N2, 40°CCambio disolvente

Extracto concentrado0.25ml

Figura 5.1. Proceso experimental para el screening de contaminantes orgánicos.


164

Utilizando la metodología anterior, ha sido posible identificar un gran número de

contaminantes. Cabe destacar la identificación de compuestos presentes en derivados

plásticos procedentes de la industria como son los compuestos fenólicos y también

compuestos OPs. El di(2-ethylhexyl) adipate ha sido encontrado en una muestra salina,

componente generalmente asociado a productos plásticos. De hecho, este compuesto ha

sido encontrado en diferentes alimentos, incluyendo alimento infantil, debido a la

migración de los componentes plásticos que envuelven a los alimentos (Petersent y

Breindahl et al., 2000). El di(2-ethylhexyl) adipate, al igual que el benzyl butyl phthalate,

encontrados en la misma muestra, son compuestos considerados como disruptores

endocrinos y su estudio es de gran importancia por ser compuestos de riesgo para la salud

humana (Ghisari et al., 2009). Recientemente, la presencia de compuestos como el

galaxolide y los OPs también ha sido reportada por otros autores señalando que se tratan

de compuestos ampliamente extendidos en el medio ambiente y por ello se debe controlar

su presencia (Schwarzbauer y Ricking et al., 2010). Por otro lado, cabe destacar que no

fueron detectados en las muestras compuestos POPs del tipo PAH, PBDE, OC o PCB

mediante el screening non-target. Entre los compuestos detectados, algunos presentan

diferentes riesgos medioambientales y en humanos debido a su toxicidad (ver Tabla 1.

Artículo científico 5).

• Non-target screening mediante GC-TOF MS.

La metodología de screening desarrollada en este estudio permitió la identificación

de contaminantes presentes en las muestras y la posibilidad de obtener su composición

elemental. La elevada exactitud de masa de los analizadores TOF junto a su excelente

sensibilidad y selectividad en el análisis de compuestos orgánicos a niveles de traza, resulta

una herramienta esencial para llevar a cabo una identificación satisfactoria. Como ejemplo,

la Figura 5.2. muestra información relevante para un posible positivo de TPhP en una

muestra salina localizada en el Delta del Ebro. En la Figura se presenta una superposición de

los iones principales, seleccionados automáticamente por el software para la identificación

del compuesto propuesto (A). El software identifica un pico cromatográfico y propone su

espectro de deconvolución que es comparado con un espectro en masa nominal de librería


165

(B, C). Los parámetros “Forward Fit” y “Reverse Fit” representan la concordancia del

espectro experimental con el de librería. Cuanto mayor sean estos valores, más semejante

es el espectro experimental del compuesto propuesto con el teórico de librería.

m/z50 100 150 200 250 300

%

0

100

6.226e+002Suggested compound: Triphenyl phosphate

326.0724

325.0637

77.0362

169.0661

168.0681

119.0853

215.0280233.0410

249.0192281.0534

m/z

%

0

100

Library: Triphenyl phosphate; C18H15O4P

326

325

77

65

5139

170

9493

168141

95128

215

171187

233

249307289

327

328

min30.600 30.800 31.000 31.200 31.400

%

0

100

9.103e+002Triphenyl phosphate

AB

C

DCompound Formula Mol. Weight Forward Fit Reverse Fit Nº CAS 326.0724 325.0637 215.0280 77.0362 65.0363

TriphenylPhosphate C18H15O4P 326 788 856 115-86-6

1.6, C18H15O4P

0.7, C18H14O4P

1.8, C12H8O2P

-2.9, C6H5

-2.8, C5H5

Figura 5.2. Información obtenida a partir del software ChromaLynx para la identificación de compuestos non-target. (A) Identificación de TPhP en una muestra salina. (B) Espectro nominal teórico del TPhP a partir de librería. (C) Espectro de deconvolución experimental para el TPhP. (D) Match de librería y errores de masas (mDa) de los principales fragmentos del espectro.

Para el caso del TPhP, estos valores son 788 y 856 respectivamente, demostrando

una gran similitud entre el espectro de deconvolución que ofrece el software comparado

con el espectro de librería. Posteriormente, se presenta información acerca de los errores

de masas de fragmentos del espectro (D, error obtenido entre m/z experimental-m/z teórica). El


166

software compara la masa experimental obtenida para un ión seleccionado con su masa

teórica (obtenida a partir de la herramienta Molecular Weight Calculator). Cuanto más bajo

sea el error de masa, la identificación es más fiable. Para el caso del TPhP mostrado

anteriormente, la identificación se pudo llevar a cabo satisfactoriamente obteniendo

errores de masa <3mDa para cinco iones principales del espectro.

El software asigna una composición elemental a cada ión m/z encontrado. Este

proceso se realiza teniendo en cuenta todas aquellas combinaciones de átomos posibles,

sin considerar si esta composición elemental tiene sentido como fragmento de la estructura

del compuesto identificado. Este proceso constituye una limitación a tener en cuenta en el

proceso de confirmación. Por ello, es importante revisar cada una de las composiciones

elementales asignadas y estudiar si estas pueden corresponder a fragmentos del

compuesto identificado.

Exclusivamente para el caso de los OPs encontrados, se llevó a cabo una

confirmación adicional con patrones de referencia. De esta manera fue posible comparar

los resultados obtenidos con los mimos obtenidos con patrones. Las desviaciones entre el

tiempo de retención y relación de Q/q fue atendida siguiendo las instrucciones de la

reglamentación vigente (Commission Decision 2002/657/EC) para evitar reportar falsos

positivos.

Previamente a nuestro estudio ya se habían publicado trabajos utilizando

analizadores TOF para el screening non-target de compuestos orgánicos de interés. García-

Reyes et al., (2007) discuten las distintas posibilidades de trabajo en modo screening con

LC-TOF MS para el análisis de residuos de plaguicidas en muestras alimentarias. El trabajo

manifiesta la excelente sensibilidad en modo de adquisición de espectro completa y la

posibilidad de realizar medidas de masa exacta trabajando con analizadores TOF para la

identificación de plaguicidas a niveles de concentración bajos. La utilidad de los

analizadores TOF ha crecido paulatinamente y se puede encontrar en bibliografía trabajos

excelentes tanto para aplicaciones con LC-QTOF MS o GC-TOF MS gracias a sus excelentes


167

prestaciones para la búsqueda de compuestos desconocidos mediante metodología non-

target (Hernández et al., 2007; Ibañez et al., 2008; Zhou et al., 2009; Guthery et al., 2010).

5.2.4. Referencias.

Čajka, T.; Hajšlová, J. 2004. Gas chromatography-high-resolution time-of-flight mass

spectrometry in pesticide residue analysis: Advantages and limitations. J.

Chromatogr. A 1058, 251-261.



http://eur-ex.europa.eu/LexUriServ/LexUriServ.do?uri=OJ:L:2002:221:0008:0036:

EN: PDF.

Dallüge, J.; Van Rijn, M.; Beens, J.; Vreuls, R.J.J.; Brinkman, U.A.Th. 2002. Comprehensive

two-dimensional gas chromatography with time-of-flight mass spectrometric

detection applied to the determination of pesticides in food extracts. J. Chromatogr.

A 965, 207-217.

Feo, M.L.; Eljarrat, E.; Barceló, D.; Barceló, D. 2010. Determination of pyrethroid

insecticides in environmental samples. Trens Anal. Chem. 29, 692-705.

Ferrer, I.; Thurman, E.M. 2003. Liquid chromatography/time-of-flight/mass spectrometry

(LC/TOF/MS) for the analysis of emerging contaminants. Trends Anal. Chem. 22, 750-

756.

García-Reyes, J.F.; Hernando, M.D.; Molina-Díaz, A.; Fernández-Alba, A.R. 2007.

Comprehensive screening of target, non-target and unknown pesticides in food by

LC-TOF-MS. Trends Anal. Chem. 26, 828-841.


168

Grimalt, S.; V.Sancho, J.; Pozo, O.J.; Hernández, F. 2010. Quantification, confirmation and

screening capability of UHPLC coupled to triple quadrupole and hybrid quadrupole

time-of-flight mass spectrometry in pesticide residue analysis. J. Mass Spectrom. 45,

421-436.

Ghisari, M.; Bonefeld-Jorgensen, E.C. 2009. Effects of plasticizers and their mixtures on

estrogen receptor and thyroid hormone functions. Toxicol. Lett. 189, 67-77.

Guthery, B.; Bassindale, T.; Bassindale, A.; Pillinger, C.T.; Morgan, G.H. 2010. Qualitative

drug analysis of hair extracts by comprehensive two-dimensional gas

chromatography/time-of-flight mass spectrometry. J. Chromatogr. A 1217, 4402-

4410.

Hernández, F.; Portolés, T.; Pitarch, E.; López, F.J. 2007. Target and nontarget screening of


chromatography/high-resolution time-of-flight mass spectrometry. Anal. Chem. 79,

9494-9504.

Hernández, F.; Portolés, T.; Pitarch, E.; López, F.J. 2009. Searching for anthropogenic


flight mass spectrometry. J. Mass Spectrom. 44, 1-11.

Howard, P.H.; Muir, D.C.G. 2010. Identifying new persistent and bioaccumulative organics

among chemicals in commerce. Environ. Sci. Technol. 44, 2277-2285.

Ibáñez, M.; Sancho, J.V.; Hernández, F.; McMillan, D.; Rao, R. 2008. Rapid non-target

screening of organic pollutants in water by ultraperformance liquid chromatography

coupled to time-of-light mass spectrometry. Trends Anal. Chem. 27, 481-489.

Muir, D.C.G.; Howard, P.H. 2006. Are there other persistent organic pollutants? A challenge

for environmental chemists. Environ. Sci. Technol. 40, 7157-7166.

Petersent, J.H.; Breindahl, T. 2000. Plasticizers in total diet samples, baby food and infant

formulae. Food Addit. Contam. 17, 133-141.


169

Portolés, T.; Pitarch, E.; López, F.J.; Sancho, J.V.; Hernández, F. 2007. Methodical approach

for the use of GC-TOF MS for screening and confirmation of organic pollutants in

environmental water. J. Mass Spectrom. 42, 1175-1185.

Schwarzbauer, J.; Ricking, M. 2010. Non-target screening analysis of river water as

compound-related base for monitoring measures. Environ. Sci. Pollut. R. 17, 934-947.

Zhou, J.-L.; Qi, L.-W.; Li, P. 2009. Herbal medicine analysis by liquid chromatography/time-

of-flight mass spectrometry. J. Chromatogr. A 1216, 7582-7594.

Capítulo 5. Determinación de OPEs mediante GC-TOF MS

171

5.3. Desarrollo de metodología analítica para la identificación de OPEs en muestras

medioambientales.


172


173


Como se ha indicado anteriormente, en la actualidad una gran cantidad de los

compuestos químicos utilizados industrialmente o asociados a materiales de diferente

naturaleza como plásticos, pinturas, productos de limpieza, etc, provocan efectos

perjudiciales sobre la salud y el medioambiente. Legislación europea reciente pretende

proporcionar un marco legal para tratar estos productos químicos con el fin de asegurar un

alto nivel de seguridad y protección para el medio ambiente

(http://echa.europa.eu/home_es.asp). Dentro de la amplia lista de compuestos

susceptibles de provocar algún efecto adverso se encuentran los POPs, que incluyen a los

PCBs, dioxinas (PCDDs), dibenzofuranos (PCDFs), OCs, PAHs, PBDEs, entre otros. También

aparecen nuevos compuestos de reciente interés como ftalatos, fosfatos retardantes de

llama, perfumes, pesticidas, biocidas, y alquilos perfluorados (Garcia-Jares et al., 2009;

Richardson et al., 2010). Sobre todos estos compuestos existe un creciente interés por su

presencia en el medio ambiente y los efectos adversos que pueden producir.

Dentro de nuestra investigación, a partir de la realización de un screening de

compuestos desconocidos en muestras de sal, se identificaron cuatro compuestos OPEs:

TCPP, TBP, TPhP y EHDPP. Estos compuestos pueden ser un claro ejemplo de

contaminantes emergentes dado que han sido frecuentemente identificados en muestras

ambientales en los últimos años (Reemtsma et al., 2008). La mayor parte de los OPEs han

sido producidos y siguen produciéndose en proporciones elevadas y son ampliamente

utilizados como retardantes de llama (Reemtsma et al., 2008), por lo que son objeto de una

creciente preocupación y hace necesario su control en alimentos y diferentes recursos

naturales.

En nuestro caso, la identificación y detección de estos compuestos en muestras

analizadas previamente nos llevó a desarrollar una metodología avanzada basada en GC-

TOF MS para la identificación de 12 compuestos OPEs en muestras acuosas, sales y

salmueras.


174

En el artículo que se presenta a continuación se muestran diferentes enfoques

analíticos utilizando el GC-TOF MS para el screening de contaminantes orgánicos. En primer

lugar se llevó a cabo la selección de los iones para la identificación de cada compuesto. Se

seleccionaron como mínimo cuatro iones por compuesto para una correcta identificación.

En segundo lugar, se realizó un análisis target de muestras seleccionadas para el análisis de

OPEs. Por último, se realizó un análisis post-target de OPEs en muestras analizadas en el

equipo con anterioridad gracias a la adquisición de espectro completo del TOF MS.


175


Investigation of organophosphate esters in fresh water, salt and brine samples by GC-TOF

MS.

J. Nácher-Mestre, R. Serrano, T. Portolés, F. Hernández.



176


177

Investigation of OPEs in fresh water, salt and brine samples by GC-TOF

MS.

Jaime Nácher-Mestre, Roque Serrano, Tania Portolés, Félix Hernández*

Research Institute for Pesticides and Water (IUPA). Avda Sos Baynat, s/n.

University Jaume I, 12071 Castellón, Spain.

Abstract

Advanced analytical methodology based on gas chromatography coupled to a high

resolution time-of-flight mass analyzer (GC-TOF MS) is proposed for investigation of

organophosphate esters (OPEs) in environmental water, marine salts and brine samples.

The target screening was carried out for 12 OPEs by evaluating the presence of up to four

m/z ions for every compound. The identification criterion was the presence of, at least, two

m/z ions at expected retention time, measured at their accurate mass, and the agreement

of the Q/qi ratio when compared to reference standards. This procedure allows the

detection and reliable identification of the compounds in the samples at very low

concentration levels (ng/g). The developed methodology was applied to commercial marine

salt, brines and different environmental water samples obtaining positive findings for

several OPEs such as tri(chloro-propyl)phosphate, tributhyl phosphate, triphenyl phosphate

among other organophosphate compounds. Thanks to the full-spectrum acquisition

performed in GC-TOF MS, it is possible to make a retrospective data evaluation without the

need of performing additional analysis. This allowed in this work to re-evaluate data

acquired in a previous GC-TOF MS non-target analysis of some marine salt and water

samples with the result of discovering several of the OPEs investigated in those samples.

Keywords: marine salt, brine, water, gas chromatography, time-of-flight mass

spectrometer, qualitative screening, organic contaminants.

* Corresponding author. Tel. +34-964-387366; e-mail address: [email protected]


178

Introduction

Nowadays, there is a rising concern about toxicity and environmental effects of compounds

that are not included in commission regulations, legislations or directives which inform

about the control levels of these called “new contaminants” in environmental matrices [1].

Organophosphorus compounds (OPs) are a family of organic compounds which are

widespread found in the environment because they have been used in different

applications such as plastizers, flame retardants, pesticides, antifouling or textiles. Several

authors have reported high levels of organophosphate esters (OPEs), a sub-family of OPs, in

environmental samples like environmental and waste water, air or sediments [2-6]. OPEs

have been recently under review due to their adverse effects as neurotoxics and

carcinogens and tend to be present in the environment at high concentrations. These

properties have increased the concern about the potential negative impact of these

substances in the environment and for this reason they were classified as “re-emerging

contaminants” [2]. Chlorinated OPEs such as tris(2-chloroethyl)phosphate (TCEP),

tri(chloro-propyl)phosphate (TCPP) and tri(dichloropropyl)phosphate (TDCP) have been

considered as carcinogenic. Moreover, non-chlorinated OPEs such trimethyl phosphate

(TMP), tributhyl phosphate (TBP), triphenyl phosphate (TPhP) and tributoxyethyl phosphate

(TBEP) have also toxic effects. A widespread report of toxicity and ecotoxicology for other

OPEs was reported by Reemtsma et al. (2008) [2].

OPEs have been monitored in different environmental samples and there is a tendency to

develop new sensitive methodologies to determine low concentrations of OPEs in the

environmental water, air, sediments, urban dust samples, and also in human biological

samples, like urine [6-11]. Chromatographic techniques, both gas chromatography (GC) and

liquid chromatography (LC), coupled to mass spectrometry (MS) are normally the

techniques of choice. Quintana et al. have recently reported wide information on the

determination and outcomes of GC and LC coupled to MS and other minor sophisticated

techniques [7].


179

GC coupled to high resolution (HR) time-of-flight mass analyzer (TOF MS) is an advanced

technique very powerful from a qualitative point of view, able to provide a highly reliable

identification of GC-amenable contaminants in different sample matrices [12]. TOF MS has

high sensitivity in full-spectrum acquisition mode and high mass resolution, which provides

high mass accuracy. These characteristics lead to high selectivity by using narrow-window

eXtracted Ion Chromatograms (nw-XICs) [13]. This analyzer allows obtaining an

extraordinary amount of chemical information in a single injection making this technique

particularly attractive for the investigation of non-target compounds and also for searching

analytes in a target way [14]. Despite its excellent potential in relevant applied fields, like

environmental pollution, food safety, toxicology or doping control analysis, HRTOF MS has

been scarcely used until now [12, 15], and it is expected it will definitely grow to be the

essential, sophisticated analytical tool for the screening of organic pollutants in

environmental samples, among other applied fields.

In a previous work, we applied a non-target screening by GC-TOF MS in sea water and

marine salts from saltworks located at the Spanish Mediterranean coast [3]. The objective

was to investigate the presence in these samples of any type of organic GC-amenable

compounds that might be relevant from an environmental point of view. Without any kind

of previous analyte selection, several non-targeted compounds such as di-

(2ethylhexyl)adipate, benzyl butyl phthalate, butylated hydroxytoluene and galaxolide,

were detected. In addition, four OPEs, the most commonly reported in the literature, were

identified in several samples. This fact, considering the relevance of these contaminants,

encouraged us to widen the number of OPEs investigated in a further research. The aim of

this work is to develop a GC-TOF MS target screening method for the identification of 12

OPEs in salt from saltworks and marine salt used in aquaculture, food and feeding

purposes. The method was applied to the analysis of several samples, including commercial

marine salts, brine samples obtained from different saltworks and also to different

environmental waters (ground and surface water).


180

Experimental

Materials and reagents

Individual organophosphate reference standards were obtained from TCI Europe

(Zwijndrecht, Belgium) and Sigma Aldrich (Madrid, Spain). TMP, TPhP, 2-Ethylhexyl Diphenyl

Phosphate (EHDPP), Tris(2-ethylhexyl)Phosphate (TEHP), TBEP, Triethyl Phosphate (TEP),

TDCP, TCEP and Tricresyl Phosphate (TCrP) were purchased from TCI Europe. TBP, Tripropyl

Phosphate (TPrP) and TCPP were purchased from Sigma-Aldrich.

Individual stock solution (1 and 10 µg/ml) were prepared by dissolving reference standards

in acetone and stored in a freezer at –20 ºC. Working solutions were prepared by diluting

stock solutions in acetone for sample fortification and in n-hexane for injection in the

chromatographic system.

HPLC-grade water was obtained from a MilliQ water purification system (Millipore Ltd.,

Bedford, MA, USA). Ultratrace quality solvents such as Acetone, Ethyl Acetate,

Dichlorometane (DCM) and n-Hexane purchased from Scharlab (Barcelona, Spain), were

used in solid phase extraction (SPE) experiments. Bond Elut cartridges C18 (500 mg)

purchased from Varian (Harbor City, CA, USA), were also used for SPE.

Sample material

Samples involved in this work were the following: Nine marine salt samples (five

commercial from supermarkets and four obtained directly from manufacturers); nine brine

samples obtained from saltworks sited in the Ebro river estuary (Mediterranean coast); six

ground water and five surface water samples (both from Spanish Mediterranean area). All

samples were stored at -20ºC until analysis.

GC instrumentation

GC system (Agilent 6890N; Agilent Palo Alto, USA) equipped with an autosampler (Agilent

7683) were coupled to a time-of-flight mass spectrometer (GCT, Waters Corporation,

Manchester, U.K.), operating in electron ionization mode (EI). GC separation was performed


181

using a fused silica HP-5MS capillary column with a length of 30 m, an internal diameter of

0.25 mm and a film thickness of 0.25 μm (J&W Scientific, Folson, CA, USA). The injector

temperature was set to 280°C. Splitless injections of 1 μL of the sample were carried out.

Helium (99.999%; Carburos Metálicos, Valencia, Spain) was used as carrier gas at a flow

rate of 1 mL/min. The interface and source temperature were set to 250°C and a solvent

delay of 4 min was selected.

The oven program in the GC system was programmed, working for target analysis of OPEs,

as follows: 60°C (1min); 8°C/ min to 300°C (2 min). The TOF MS was operating at 1

spectrum/s, acquisition rate over the mass range m/z 50-650, using a multichannel plate

voltage of 2850 V. TOF MS resolution was approximately 7000 (FWHM). Heptacosa

standard, used for the daily mass calibration and as a lock mass, was injected via syringe in

the reference reservoir at 30 °C for this purpose; the m/z ion monitored was 218.9856. The

application manager TargetLynx was used to process the qualitative data obtained from

standards and from sample analysis.

Analytical procedure

The analytical procedure applied was based on the method previously developed by Pitarch

et al. (2007) [16] for the determination of priority organic pollutants in water by GC-

MS/MS, with some modifications in the pre-concentration step. Approximately 62.5g of salt

were dissolved in 250 mL of water. In the analysis of brines and environmental waters, 250

mL of sample were taken. In all cases, the filtered solution was passed through the C18 SPE

cartridge, previously conditioned by passing 6 mL methanol, 6 mL ethyl acetate:DCM

(50:50), 6 mL methanol and 6 mL water avoiding dryness. After loading the sample (250mL),

cartridges were washed with 3 mL water. Then, the cartridge was dried by passing air, using

vacuum for at least 15 min, and the elution was performed by passing 5 mL ethyl

acetate:DCM (50:50). Finally, the extract collected was evaporated under a gentle nitrogen

stream at 40°C and redisolved in 0.25 mL n-hexane and analyzed by GC-TOF MS.


182

GC-TOF MS target screening

Full spectrum acquisition data generated by TOF MS were treated using an automated

processing method. Briefly, the detection and identification of target OPEs was carried out

by obtaining between 2 and 4 nw-XICs (mass window 0.02 Da) at selected m/z exact masses

for every compound (Table 1). The presence of at least two ions at expected retention time

measured at their accurate masses was required together to the attainment of the Q/q

ratio within specified tolerances. This methodology has been previously applied by our

group and further details in development of automatic processing method can be found in

the literature [12-14].

Target analytical methodology evaluation

Prior sample analysis, this methodology was tested in marine salts that were spiked with

target OPEs at different levels. For this purpose, five marine salt samples were analyzed

using the developed procedure to identify the presence of OPEs. None of the samples was a

true blank, as all the samples analyzed contained one or more OPE. Therefore, we selected

one of these samples (the less contaminated) for spiking.

The evaluation of the overall procedure, including sample treatment and GC-TOF MS

measurement was carried out by analysis of 5 replicates of the marine salt sample spiked at

three different concentration levels: 0.4, 2 and 4 ng/g of each of the 12 OPEs selected. As

the salt was dissolved in 250 mL of HPLC-grade water, these levels would correspond to 0.1,

0.5 and 1 µg/L in brines and water samples. The evaluation of the overall procedure also

involved the analysis of two replicates of the “blank” sample used for spiking and two

method blanks to assure that no contamination was introduced in the procedure by the

routine analysis.

To consider a finding of OPE as positive, the criteria established was the presence of at least

two m/z ions at expected retention time, measured at their accurate mass, and the

agreement of their Q/q ratios within specific tolerances. The confirmation of the identity


183

was assumed by taking into account of the European Commission Decision (2002/657/CE)

[17].

The limit of identification (LOI) was established as the lowest concentration level tested at

which all the five replicates were positive, accordingly with the identification criterion

indicated above.

Results and discussion

Qualitative target investigation of OPEs

The analytical strategy developed in this paper was focused on the detection and reliable

identification of target OPEs in environmental samples more than on quantification, as the

latest can be successfully made by using the most commonly applied techniques of GC-

NPD, GC-MS and LC-MS/MS [7].

After a non-target GC-TOF MS analysis previously applied to marine salts [3], we discovered

the presence in the samples of up to 4 OPEs among other relevant contaminants.

Considering the interest of investigating OPEs in this type of samples, we extended the

research to a higher number of OPEs by developing a GC-TOF MS target screening focused

on 12 relevant OPEs. The acquisition of reference standards allowed as to test the

qualitative method in spiked samples and to perform the unequivocal confirmation of the

identity of these compounds in the samples analyzed.

The use of GC-TOF MS allows performing a retrospective analysis of the full spectrum

acquisition data acquired. Based on this, we have explored in this paper the potential of GC-

TOF MS to perform a retrospective analysis of previous data obtained in a non-target way

[3]. Data re-evaluation has been made in a post-target way, i.e. searching for selected

analytes after MS data acquisition without the need of any additional sample injection into

the GC-TOF MS [12, 18]. This possibility is not accessible using single- or triple-quadrupole

analyzers in SIM or SRM mode, respectively, or using ion trap analyzers in the MS/MS mode

where the selection of analytes must be done before the acquisition, as only analite-specific

information is acquired.


184

The target screening processing method was developed by injecting reference standards in

solvent. In general, the structures suggested for the main fragments from EI-TOF MS

spectra of the compounds, were in accordance to those available in the literature [19, 20]

(see Table 1). Many of the main m/z ions are produced by the rearrangement of more than

one hydrogen atom, giving an even-electron ion as the stable product. In the EI mode, alkyl

phosphate esters tend to lose alkyl groups by McLafferty rearrangements [21]. Basically,

this means that alkyl groups are sufficiently long to donate two hydrogens to the phosphate

interior structure during the fragmentation process. As an example, tri-alkylated phosphate

esters will afford m/z 98.9847 as the base ion [H4O4P+].

Evaluation of the target methodology

As indicated in section 2.6., the GC-TOF MS screening methodology was tested in marine

salt spiked at 0.4, 2 and 4 ng/g for each OPE selected. The LOI was established as the lowest

concentration level tested for which the identification criterion was achieved in the five

replicates tested (Table 2). Consequently, a compound was considered as satisfactory

identified at a certain concentration level, only when the five replicates spiked at that level

were positive by the accomplishment of the identification criteria established [22].

As Table 2 shows, several OPEs were identified in a reliable way at the lowest fortification

level tested. Thus, the compounds TEP, TPrP, TCEP, TDCP and TPhP achieved the

established identification criteria at 0.4 ng/g. The LOI for TBEP was found to be 2 ng/g in

salt samples, and the remaining OPES, TMP, EHDPP, TEHP and TCrP could only be correctly

identified at the highest concentration level assayed (4ng/g). It is worth to mention that

TBP and TCPP could be correctly identified at the lowest level, but no LOI could be

proposed since they were present in the “blank” sample selected for this evaluation. Table

2 also shows the equivalence of LOI in water samples.

Application to marine salt, brine and water samples. Retrospective analysis

The methodology developed in this paper was applied to the screening of OPEs in

commercial marine salt and brine samples. Table 3 shows the OPEs identified using the


185

methodology proposed. With regards to commercial marine salts, six OPEs were identified

(TEP, TPrP, TBP, TCEP, TCPP and EHDPP), three of them (TEP, TPrP and TCEP) none

previously detected using a non-target screening for the four marine salts from saltworks

[3]. Moreover, three positive findings for TBP and one for TCEP were also found in brine

samples. TCPP was present in all samples investigated. As an illustrative example, Figure 1

shows the positive finding of TBP in one commercial marine salt using the target screening

developed. In this case, 5 characteristic ions at expected retention time in the nw-XIC were

observed. The attainment of all Q/qi ratios (in this case 4) criteria within specified

tolerances led to the unequivocal confirmation of the identity of the compounds selected.

The accurate mass spectrum of the sample peak is shown together with mass errors for the

four ions (the lowest fragment was not present in the spectrum). These mass errors were

below 3mDa. In addition, chemical structures for the most abundant EI fragment ions were

suggested based on the elemental compositions proposed for those ions accordingly to the

accurate mass measurements given by the instrument in the target methodology applied

(see Table 1). All structures proposed for the fragments were compatible with the chemical

structure of TBP, making the identification still more reliable.

TOF MS allows an excellent advantage in order to reprocess data files without a new GC-

TOF MS running due to its full acquisition mode. In this sense, the developed methodology

was applied in order to investigate the presence of other OPEs in the four salt samples

previously analyzed, where four OPEs had been previously detected (TBP, TCPP, TPhP and

EHDPP) [3]. These four OPEs detected in the first study were also identified with the new

strategy but no more OPEs from the target list were found in the four salt samples

satisfying our confirmation criteria. Besides the salts, 6 ground waters and 5 surface waters

were also retrospectively analyzed. TCPP was detected in all these waters and TCEP was

identified in only one surface water.

As an illustrative example, Figure 2 shows nw-XICs for the OPEs found in the waters. This

figure shows the positive findings of TCPP, TCEP and TCPP in ground water and surface

water respectively. Five characteristic ions at expected retention time in the nw-XIC were

also observed. The attainment of all (in these case 4) Q/qi ratios criteria within specified


186

tolerances led to the unequivocal confirmation of these compounds. The qualitative

screening to these samples shows that the OPEs identified were at least over the LOI

proposed (Table 2). The presence of these compounds in these samples was in accordance

with several authors who found TCPP and other OPEs in environmental samples such

wastewater, surface water and sub-surface water [5].

Conclusions

A rapid sensitive screening with little sample manipulation is described in this work for

detection and identification of OPEs in different type of samples, based on the use of SPE

pre-concentration step and GC-TOF MS analysis. The screening method has been evaluated

in qualitative terms at different concentration levels for a total of 12 OPEs allowing the

study of these compounds in samples frequently used for aquaculture and alimentation

purposes like marine salts. For the majority of the studied analytes, LOIs were very low

(from 0.4 ng/g to 4 ng/g in salts and from 0.1 ng/mL to 1 ng/mL in waters) except for TBP

and TCPP which no LOI could be proposed due to these compounds were present in the

blank.

The screening methodology has been applied to investigate the presence of 12 OPEs in

different samples including marine salts from saltworks and commercial salts, brine, ground

water and surface water. Some of these samples were simply subjected to a retrospective

data evaluation without the need of performance an additional analysis. This was possible

because those samples were previously analyzed by GC-TOF MS and full-spectrum accurate

mass data remained available for re-evaluation, making the searching of OPEs feasible in a

post-target way. The developed methodology allowed the reliable identification of several

OPEs such as TBP, TCPP and EHDPP in marine salts, together with TPhP in salts from

saltworks, and other OPEs, such as TEP, TPrP and TCEP in commercial salt samples used for

human nutrition and alimentation purposes. All samples monitored, salts, brines and

waters, were positive for TCPP. In addition to TCPP, TBP was identified in three brines and

TCEP was identified in one brine sample and in one surface water sample.


187

Acknowledgements

The authors are grateful with Serveis Centrals d´Instrumentació Científica (SCIC) of

University Jaume I for using the GC-TOF MS. The authors acknowledge the financial support

of Generalitat Valenciana, as research group of excellence PROMETEO/2009/054.

References.

[1] Richardson SD (2008) Environmental mass spectrometry: Emerging contaminants and

current issues. Anal Chem 80:4373-4402

[2] Reemtsma T, García-López M, Rodríguez I, Quintana JB, Rodil R (2008)

Organophosphorus flame retardants and plasticizers in water and air I. Occurrence and

fate. TrAC – Trend Anal. Chem 27: 727-737

[3] Serrano R, Nácher-Mestre J, Portolés J, Amat F, Hernández F (2010) Non-target

screening of organic contaminants in marine salts by gas chromatography coupled to high-

resolution time-of-flight mass spectrometry. Submitted to Talanta.

[4] Martínez-Carballo E, González-Barreiro C, Sitka A, Scharf S, Gans O (2007)

Determination of selected organophosphate esters in the aquatic environment of Austria.

Sci Total Environ 388: 290-299

[5] Marklund A, Andersson B, Haglund P (2005) Organophosphorus flame retardants and

plasticizers in Swedish sewage treatment plants. Environ Sci Technol 39:7423-7429

[6] García-López M, Rodríguez I, Cela R, Kroening KK, Caruso JA (2009) Determination of

organophosphate flame retardants and plasticizers in sediment samples using microwave-

assisted extraction and gas chromatography with inductively coupled plasma mass

spectrometry. Talanta 79:824-829

[7] Quintana JB, Rodil R, Reemtsma T, García-López M, Rodríguez I (2008)

Organophosphorus flame retardants and plasticizers in water and air II. Analytical

methodology. TrAC – Trend Anal Chem 27:904-915

[8] Quintana JB, Rodil R, López-Mahía P, Muniategui-Lorenzo S, Prada-Rodríguez D (2007)

Optimisation of a selective method for the determination of organophosphorous triesters

in outdoor particulate samples by pressurised liquid extraction and large-volume injection

gas chromatography-positive chemical ionisation-tandem mass spectrometry. Anal Bioanal

Chem 388:1283-1293

[9] García-López M, Rodríguez I, Cela R (2009) Pressurized liquid extraction of

organophosphate triesters from sediment samples using aqueous solutions. J Chromatogr

A, 1216:6986-6993


188

[10] Schindler BK, Förster K, Angerer J (2009) Determination of human urinary

organophosphate flame retardant metabolites by solid-phase extraction and gas

chromatography-tandem mass spectrometry. J Chromatogr B 877:375-381

[11] García-López M, Rodríguez I, Cela R (2010) Mixed-mode solid-phase extraction

followed by liquid chromatography-tandem mass spectrometry for the determination of tri-

and di-substituted organophosphorus species in water samples. J Cromatogr A 1217:1476-

1484

[12] Hernández F, Portolés T, Pitarch E, Lopez FJ (2011) Gas chromatography coupled to

high-resolution time-of-flight mass spectrometry to analyze trace-level organic compounds

in environment, food safety and toxicology. TRAC-Trend Anal Chem. DOI

10.1016/j.trac.2010.11.007

[13] Portolés T, Pitarch E, López FJ, Sancho JV, Hernández F (2007) Methodical approach for

the use of GC-TOF MS for screening and confirmation of organic pollutants in

environmental water. J Mass Spectrom 42:1175-1185

[14] Hernández F, Portolés T, Pitarch E, López FJ (2007) Target and nontarget screening of


chromatography/high-resolution time-of-flight mass spectrometry. Anal Chem 79:9494-

9504

[15] Peters RJB, Oosterink JE, Stolker AAM, Georgakopoulos C, Nielen MWF (2010) Generic

sample preparation combined with high-resolution liquid chromatography-time-of-flight

mass spectrometry for unification of urine screening in doping-control laboratorios. Anal

Bioanal Chem 396:2583-2598

[16] Pitarch E, Medina C, Portolés T, López FJ, Hernández F (2007) Determination of priority

organic micro-pollutants in water by gas chromatography coupled to triple quadrupole

mass spectrometry. Anal Chim Acta 583:246-258

[17] Commission Decision 2002/657/EC implementing Council Directive 96/23/EC

concerning the performance of analytical methods and the interpretation of results.

Available: eur-


[18] Hernández F, Pozo OJ, Sancho JV, López FJ, Marín JM, Ibáñez M (2005) Strategies for

quantification and confirmation of multi-class polar pesticides and transformation products

in water by LC-MS2 using triple quadrupole and hybrid quadrupole time-of-flight analyzers.

TrAC – Trend Anal Chem 24:596-612

[19] Rodil R, Quintana JB, Reemtsma T (2005) Liquid chromatography-tandem mass

spectrometry determination of nonionic organophosphorus flame retardants and

plasticizers in wastewater samples. Anal Chem 77:3083-3089


189

[20] Ellis J, Shah M, Kubachka KM, Caruso JA (2007) Determination of organophosphorus

fire retardants and plasticizers in wastewater samples using MAE-SPME with GC-ICPMS and

GC-TOFMS detection. J Environ Monitor 9:1329-1336

[21] McLafferty FW (1959) Mass spectrometric analysis: Molecular rearrangements Anal

Chem 31:82-87

[22] Portolés T, Pitarch E, López FJ, Hernández F (2011) Development and validation of a

rapid and wide-scope qualitative screening method for detection and identification of

organic pollutants in natural water and wastewater by gas chromatography time-of-flight

mass spectrometry. J Chromatogr A 1218:303-315

Figure Captions.

Figure 1. Extracted-ion chromatograms (mass window 0.02 Da) showing a positive finding

of TBP in commercial marine salt. Experimental EI accurate mass spectrum. Chemical

structures proposed for the most abundant EI fragment ions. St: reference standard; S:

sample; Q: qualitative ion; q: confirmation ion; �: Q/q ratio within tolerance limits.

Figure 2. Extracted ion chromatogram at different m/z (mass window 0.02 Da) for TCPP in

ground water and, TCEP and TCPP in surface waters. St: reference standard; S: sample; Q:

qualitative ion; q: confirmation ion; �: Q/q ratio within tolerance limits.


190

Table1. m/z ions selected for the identification of OPEs.

Compound

CAS tR Formula MW Ion1 m/z1 Ion2 m/z2 Ion3 m/z3 Ion 4 m/z4 Ion5 m/z5

Trimethyl phosphate TMP

512-56-1 4.17 C3H9O4P 140.0238 C3H9O4P 140.0238 C2H7O3P 110.0133 C2H6O3P 109.0055 CH5O2P 80.0027 CH4O2P 78.9949

Triethyl phosphate TEP

78-40-0 7.38 C6H15O4P 182.0708 C4H12O4P 155.0473 C3H8O4P 139.0160 C2H8O4P 127.0160 C2H6O3P 109.0055 H4O4P 98.9847

Tripropyl phosphate TPrP

513-08-6 11.83 C9H21O4P 224.1177 H4O4P 98.9847 C6H16O4P 183.0786 C4H10O4P 153.0317 C3H10O4P 141.0317 C3H8O3P 123.0211

Tributyl phosphate TBP

126-73-8 15.85 C12H27O4P 266.1647 H4O4P 98.9847 C8H20O4P 211.1099 C4H12O4P 155.0473 C2H6O4P 125.0004 C4H8 56.0626

Tris-(2-chloroethyl)phosphate TCEP

115-96-8 17.43

C6H12Cl3O4P 283.9539 C4H8Cl2O3P 204.9588 C6H12Cl2O4P 248.9850 H4O4P 98.9847 C2H5ClO3P 142.9665 C2H4Cl 63.0002

Tris(1-chloro-2-propyl) phosphate TCPP

13674-84-5 17.95 C9H18Cl3O4P 326.0008 H4O4P 98.9847 C5H11ClO4P 201.0084 C3H7ClO3P 156.9821 C3H9ClO4P 174.9927 C8H16Cl2O4P 277.0163

Tris(1,3-dichloro-2-propyl)phosphate TDCP

13674-87-8 24.19 C9H15Cl6O4P 427.8839 C3H6Cl2O3P 190.9432 C3H8Cl2O4P 208.9537 C3H5ClO3P 154.9665 C3H4Cl2 109.9690 C3H4Cl 75.0002

Triphenyl phosphate TPhP

115-86-6 24.85 C18H15O4P 326.0708 C18H15O4P 326.0707 C12H8O2P 215.0262 C6H6O 94.0419 C6H5 77.0391 - -

Tri(butoxyethyl)phosphate TBEP

78-51-3 24.94 C18H39O7P 398.2433 C5H9O 85.0653 C6H6O 94.0419 C2H6O4P 125.0004 C6H12O 100.0888 - -

2-Ethylhexyl diphenyl phosphate EHDPP

1241-94-7 25.14 C20H27O4P 362.1647 C12H12O4P 251.0473 C12H11O4P 250.0395 C12H10O4P 249.0317 C12H10O 170.0732 - -

Tris(ethylhexyl)phosphate TEHP

78-42-2 25.52 C24H51O4P 434.3525 H4O4P 98.9847 C8H16 112.1252 C8H17 113.1330 C6H11 83.0861 C5H10 70.0783

Tricresyl phosphate TCrP

1330-78-5 27.29 C21H21O4P 368.1177 C21H21O4P 368.1177 C21H20O4P 367.1099 C7H8O 108.0575 C7H7O 107.0497 - -


191

Table 2. Evaluation of the procedure when applied to spiked samples (n=5). Limits of identification.

LOI

Marine Salt

(ng/g)

LOI

Water a

(µg/L)

Compound Nº CAS Formula

Trimethyl phosphate TMP 512-56-1 C3H9O4P 4 1

Triethyl phosphate TEP 78-40-0 C6H15O4P 0.4 0.1

Tripropyl phosphate TPrP 513-08-6 C9H21O4P 0.4 0.1

Tributyl phosphate TBP 126-73-8 C12H27PO4 (b) (c)

Tris-(2-chloroethyl)phosphate TCEP 115-96-8 C6H12Cl3O4P 0.4 0.1

Tris(1-chloro-2-propyl) phosphate TCPP 13674-84-5 C9H18Cl3O4P (b) (c)

Tris(1,3-dichloro-2-propyl)phosphate TDCP 13674-87-8 C9H15Cl6O4P 0.4 0.1

Triphenyl phosphate TPhP 115-86-6 C18H15O4P 0.4 0.1

Tri(butoxyethyl)phosphate TBEP 78-51-3 C18H39O7P 2 0.5

2-Ethylhexyl diphenyl phosphate (Octizicer) EHDPP 1241-94-7 C20H27O4P 4 1

Tris(2-ethylhexyl)phosphate TEHP 78-42-2 C24H51O4P 4 1

Tricresyl phosphate TCrP 1330-78-5 C21H21O4P 4 1

a: Values that would correspond to water samples considering the treatment applied to salt samples.

b: Compounds identified at 0.4 ng/g level, although they were present in the “blank” sample used in the experiments.

c: Compounds identified at a level that would correspond to 0.1 µg/L, although they were also present in the “blank” sample used in the experiments.


192

Table 3. Identification of OPEs in different type of samples using the proposed methodology.

Number of positives / number of samples analyzed

New samples analyzed

Retrospective analysis of samples

previously injected into GC-TOF MS

Compound

5

commercial

marine salts

9

brine water

4

marine salts

from

saltworks

6

ground

waters

5

surface

waters


0/5

0/9

0/4

0/6

0/5


0/5

0/9

0/4

0/6

0/5

Triethyl phosphate TEP

4/5

0/9

0/4

0/6

0/5

Tripropyl phosphate TPrP

1/5

0/9

0/4

0/6

0/5

Tributyl phosphate TBP

5/5

3/9

4/4

0/6

0/5

Tris-(2-chloroethyl)phosphate TCEP

1/5

1/9

0/4

0/6

1/5

Tris(1-chloro-2-propyl) phosphate TCPP

5/5

9/9

4/4

6/6

5/5

Tris(1,3-dichloro-2-propyl)phosphate TDCP

0/5

0/9

0/4

0/6

0/5

Triphenyl phosphate TPhP

0/5

0/9

2/4

0/6

0/5

Tri(butoxyethyl)phosphate TBEP

0/5

0/9

0/4

0/6

0/5

2-Ethylhexyl diphenyl phosphate

(Octizicer) EHDPP

2/5

0/9

2/4

0/6

0/5

Tris(2-ethylhexyl)phosphate TEHP

0/5

0/9

0/4

0/6

0/5

Tricresyl phosphate TCrP 0/5 0/9 0/4 0/6 0/5


193

S12

m/z60 80 100 120 140 160 180 200 220 240 260 280

%

0

100

MP317 834 (17.902) Cm (834:839-(787:828+853:870)) TOF MS EI+ 7.22e398.9817

82.9856

155.0462

124.9986

112.0052

211.1089

167.0577 205.1641 213.1519249.1904257.1913

S12

Time17.00 18.00 19.00

%

0

100

17.00 18.00 19.00

%

0

100

17.00 18.00 19.00

%

0

100

17.00 18.00 19.00

%

0

100

17.00 18.00 19.00

%

0

100

MP317 TOF MS EI+ 98.984 0.02Da

1.93e3Area

17.88;255

MP317 TOF MS EI+ 211.109 0.02Da

191Area

17.88;25

MP317 TOF MS EI+ 155.047 0.02Da

467Area

17.88;61

MP317 TOF MS EI+ 124.999 0.02Da

172Area

17.90;22

MP317 TOF MS EI+ 56.062 0.02Da

579Area

17.90;64

O

P O+

O

O

H

CH3

CH3

HC8H20O4P

O

P O+

O

OH

H

CH3

H

C4H12O4P

C2H6O4P

H O+

P O

O

OH

O P O+

OH

OH

H

H

H4O4P

-1 mDa

-1.1 mDa

-1.8 mDa

-3 mDa

Q/q (St)= 9.71Q/q (S)= 10.25% �

Q/q (St)= 4Q/q (S)= 4.185% �

Q/q (St)= 11.33Q/q (S)= 10.597% �

Q/q (St)= 5.25Q/q (S)= 3.9824% �

CH2+

CH2

C4H8

? mDa

Figure 1.


194

Surfacewater TCPP

Time20.00 21.00 22.00

%

0

100

20.00 21.00 22.00

%

0

100

20.00 21.00 22.00

%

0

100

20.00 21.00 22.00

%

0

100

20.00 21.00 22.00

%

0

10098.984 0.02Da21.03;83

201.008 0.02Da21.03;50

156.982 0.02Da21.03;40

174.992 0.02Da21.03;19

277.016 0.02Da21.03;18

Groundwater TCPP

Time19.00 20.00 21.00

%

0

100

19.00 20.00 21.00

%

0

100

19.00 20.00 21.00

%

0

100

19.00 20.00 21.00

%

0

100

19.00 20.00 21.00

%

0

10098.984 0.02Da20.22;4

201.008 0.02Da20.22;3

156.982 0.02Da20.22;2

174.992 0.02Da20.22;1

277.016 0.02Da20.22;1

Q/q (St)= 1.45Q/q (S)= 1.338% �

Q/q (St)= 1.88Q/q (S)= 26% �

Q/q (St)= 4.46Q/q (S)= 410% �

Q/q (St)= 4.42Q/q (S)= 410% �

Surfacewater TCEP

Time20.50 21.00

%

0

100

20.50 21.00

%

0

100

20.50 21.00

%

0

100

20.50 21.00

%

0

100

20.50 21.00

%

0

100204.958 0.02Da20.59;12

248.985 0.02Da20.59;10

98.984 0.02Da20.59;5

142.966 0.02Da20.59;9

63. 0.02Da20.59;10

Q/q (St)= 1.45Q/q (S)= 1.338% �

Q/q (St)= 1.3Q/q (S)= 1.28% �

Q/q (St)= 2.7Q/q (S)= 2.411% �

Q/q (St)= 1.1Q/q (S)= 1.210% �

Q/q (St)= 1.45Q/q (S)= 1.6615% �

Q/q (St)= 1.88Q/q (S)= 2.0710% �

Q/q (St)= 4.46Q/q (S)= 4.362% �

Q/q (St)= 4.42Q/q (S)=4.61 4% �

Figure 2.


195

5.3.3. Discusión de resultados.

• Metodología screening de OPEs.

Para llevar a cabo una correcta identificación de la presencia de OPEs en las muestras

mediante GC-TOF MS, se seleccionaron al menos cuatro iones característicos por

compuesto. Se consideró la masa exacta de dos iones por compuesto como criterio mínimo

para alcanzar una buena identificación atendiendo a los criterios de identificación

propuestos por la legislación europea (Commission Decision 2002/657/EC). El tiempo de

retención además de las relaciones Q/q fueron comparadas entre muestras y patrones

certificados para cada compuesto con el fin de asegurar una correcta identificación

(Nácher-Mestre et al., 2009).

La masa teórica experimental de los OPEs seleccionados fue obtenida a partir de la

adquisición del espectro completo mediante GC-TOF MS al inyectar en el equipo patrones

certificados de cada compuesto. De esta forma ha sido posible obtener picos

cromatográficos para una masa exacta determinada y poder confirmar la presencia

inequívoca de OPEs en las muestras analizadas. La exactitud de masa resulta ser un

parámetro crítico para la identificación de compuestos puesto que permite la confirmación

de la identidad de los OPEs con errores de masa normalmente bajos (Ferrer et al., 2006).

Así pues, a cada relación m/z le corresponde una composición elemental lo que limita

extraordinariamente la posibilidad de confusión con otro compuesto. Adquiriendo más de

un ión identificativo por compuesto mediante GC-TOF MS se obtiene una correcta

identificación y exactitud de masa con hasta 3 ó 4 puntos de identificación (IPs), requeridos

para compuestos autorizados o prohibidos, respectivamente (Commission Decision

2002/657/EC).

La utilización de GC-TOF MS ha resultado ser una herramienta muy útil para la

identificación de OPEs en matrices medioambientales y de consumo humano. Los

analizadores TOF han sido también utilizados para obtener una confirmación adicional de la

presencia de contaminantes gracias a su elevada resolución y exactitud de masa (Nácher-

Mestre et al., 2009; Pitarch et al., 2010). Recientemente, Ellis et al. (2007) ha desarrollado


196

una metodología para la determinación de OPEs mediante GC-MS con fuente de plasma de

acoplamiento inductivo (GC-ICP MS) y con la utilización de GC-TOF MS mostrando así un

gran poder de identificación de los analitos.

• Evaluación de la metodología.

El método para la identificación de OPEs fue evaluado con el fin de realizar una

estimación real de la identificación de OPEs a concentraciones por encima de los LOI

establecidos. Los resultados fueron satisfactorios pudiendo proponer LOIs a niveles de ng/g

para sales marinas y a niveles de µg/L para las aguas, para todos los OPEs seleccionados.

Únicamente para los compuestos TBP y TCPP no se pudieron establecer LOIs puesto que

estos analitos estaban presentes en la matriz “blanco”. El criterio para reportar un positivo

fue la identificación de al menos dos fragmentos iónicos a su masa exacta sin que

presentaran desviaciones apreciables en la relación Q/q (Commission Decision

2002/657/EC). Así se obtuvo un alto grado de confirmación para la presencia de OPEs en

muestras acuosas y muestras con elevado contenido salino a niveles de concentración

extraordinariamente bajos.

• Screening post-target de OPEs en muestras reales. Análisis retrospectivo.

Gracias a que el analizador TOF presenta una elevada capacidad resolutiva y también

gracias a su modo de adquisición en full scan con alta sensibilidad, es posible realizar una

búsqueda de un analito específico con posterioridad a la adquisición de datos por MS,

siempre conociendo sus fragmentos iónicos teóricos. Así, contaminantes orgánicos

diferentes a los seleccionados en muestras previamente adquiridas en analizadores TOF

pueden ser identificados a posteriori (Rubio et al., 2009). En nuestro caso, la información

espectral completa adquirida por los analizadores TOF nos permitió identificar analitos

seleccionados en modo post-target. Así pues, fue posible llevar a cabo el análisis de OPEs

en muestras previamente analizadas mediante GC-TOF MS, muestras en las que no se había

llevado a cabo el estudio de la presencia de OPEs previamente. Igualmente, una de las

ventajas fue el hecho de no ser necesario realizar un nuevo tratamiento a las muestras

puesto que el analizador TOF adquiere todo el espectro de masas y es posible identificar la


197

presencia de cualquier compuesto ionizable en un sistema de GC a posteriori. El procesado

de la información mediante el software TargetLynx para los compuestos seleccionados,

permitió su identificación en diferentes muestras.

Previamente a nuestro trabajo, Portolés et al. (2007) realizó una aplicación con TOF

MS que permitió la investigación de la presencia de varios analitos en muestras acuosas

mediante análisis post-target mostrando la gran ventaja de no reanalizar las muestras de

nuevo. Esta posibilidad de análisis no es posible con otros analizadores del tipo Q o QqQ

trabajando en modos SIM o SRM o trabajando con analizadores IT en modo MS/MS puesto

que la selección de los iones se realiza antes de la adquisición en MS. Recientemente,

Grimalt et al. (2010) realizó una comparativa de tres analizadores de MS, QqQ, TOF y QTOF

en la que el analizador TOF mostró mayores ventajas que los demás para el screening por

su mejor sensibilidad trabajando en modo de adquisición de espectro completo y medidas

de masa exacta.

5.3.4. Referencias.



eur-lex. europa.eu/pri/en/oj/dat/2002/l_221/l_22120020817en000-80036.pdf.

Ellis, J.; Shah, M.; Kubachka, K.M.; Caruso, J.A. 2007. Determination of organophosphorus

fire retardants and plasticizers in wastewater samples using MAE-SPME with GC-

ICPMS and GC-TOFMS detection. J. Environ. Monitor. 9, 1329-1336.

Ferrer, I.; Fernandez-Alba, A.; Zweigenbaum, J.A.; Thurman, E.M. 2006. Exact-mass library

for pesticides using a molecular-feature database. Rapid Commun. Mass Spectrom.

20, 3659-3668.

Garcia-Jares, C.; Regueiro, J.; Barro, R.; Dagnac, T.; Llompart, M. 2009. Analysis of industrial

contaminants in indoor air. Part 2. Emergent contaminants and pesticides. J.

Chromatogr. A 1216, 567-597.


198

Grimalt, S.; V.Sancho, J.; Pozo, O.J.; Hernández, F.E. 2010. Quantification, confirmation and

screening capability of UHPLC coupled to triple quadrupole and hybrid quadrupole

time-of-flightmass spectrometry in pesticide residue analysis. J. Mass Spectrom. 45,

421-436.

McLafferty, F.W. 1959. Mass spectrometric analysis: Molecular rearrangements. Anal.

Chem. 31, 82-87.

Nácher-Mestre, J.; Serrano, R.; Portolés, T.; Hernández, F.; Benedito-Palos, L.; Pérez-

Sánchez, J., 2009. A reliable analytical approach based on gas chromatography

coupled to triple quadrupole and time-of-flight mass analyzers for the determination

and confirmation of polycyclic aromatic hydrocarbons in complex matrices from

aquaculture activities. Rapid Commun. Mass Spectrom. 23, 2075-2086.

Pitarch, E.; Portolés, T.; Marín, J.M.; Ibáñez, M.; Albarrán, F.; Hernández, F. 2010. Analytical

strategy based on the use of liquid chromatography and gas chromatography with

triple-quadrupole and time-of-flight MS analyzers for investigating organic

contaminants in wastewater. Anal. Bioanal. Chem. Article in Press.

Portolés, T.; Pitarch, E.; López, F.J.; Sancho, J.V.; Hernández, F. 2007. Methodical approach

for the use of GC-TOF MS for screening and confirmation of organic pollutants in

environmental water. J. Mass Spectrom. 42, 1175-1185.

Reemtsma, T., García-López, M., Rodríguez, I., Quintana, J.B., Rodil, R. 2008.

Organophosphorus flame retardants and plasticizers in water and air I. Occurrence

and fate. Trends Anal. Chem. 27, 727-737.

Rubio, S.; Pérez-Bendito, D. 2009. Recent advances in environmental analysis. Anal. Chem.

81, 4601-4622.

CAPÍTULO 6

ESTUDIO DE LA BIOACUMULACIÓN DE CONTAMINANTES ORGÁNICOS PERSISTENTES EN

PRODUCTOS DE LA ACUICULTURA MARINA.


201

6.1. Introducción


202


203

En este capítulo de la Tesis Doctoral se presentan dos estudios sobre la

bioacumulación de contaminantes orgánicos en productos de la acuicultura marina.

En los últimos tiempos, la creciente demanda de aceites de pescado por parte de la

acuicultura, unido a un decrecimiento de las capturas en las pesquerías, ha puesto en

peligro la sostenibilidad de esta industria. Esto ha hecho necesaria la búsqueda de fuentes

alternativas de lípidos para los piensos utilizados en la cría de peces.

Actualmente se estudia la posibilidad de sustituir el aceite de pescado presente en

las formulaciones de piensos, al menos en parte, por aceites de origen vegetal. Esto hace

necesario conocer las consecuencias que esta sustitución puede tener en la calidad del

producto final, como por ejemplo en el perfil de ácidos grasos de los peces cultivados con

estos piensos o la carga de contaminantes en el producto final.

En las instalaciones del Instituto de Acuicultura de Torre la Sal (IATS, CSIC), en el

marco del proyecto europeo “Sustainable Aquafeeds to Maximise the Health Benefits of

Farmed Fish for Consumers” (Codigo n° 016249-2), se llevó a cabo un experimento que

consistió en el cultivo de doradas (Sparus aurata L.) a lo largo de un ciclo de crecimiento

completo, utilizando dietas con composiciones que diferían en la proporción de aceites de

pescado, sustituido por aceites de origen vegetal. A partir de este experimento, se

estudiaron paralelamente parámetros de crecimiento, perfiles de ácidos grasos y otros

indicadores bioquímicos, junto con el estudio de la presencia de POPs en los diferentes

productos utilizados o producidos durante todo el proceso de cultivo.

Concretamente se estudió la presencia de PAHs, OCPs, PBDEs y PCBs en los piensos

que fueron suministrados a los peces durante el experimento, en los ingredientes utilizados

en la formación de estos, y en los filetes de los peces a lo largo de todo el proceso de

engorde.

El estudio de la bioacumulación de PAHs y OCs dio lugar a los artículos científicos que

se presentan y comentan a continuación. La presencia de PBDEs en las muestras analizadas

presentó valores de concentración por debajo del LOQ en la mayoría de los casos, no

observándose bioacumulación de estos compuestos a lo largo del proceso de engorde

(Artículo científico 6).


204

Capítulo 6. Bioacumulación de PAHs en doradas procedentes de la acuicultura

205

6.2. Estudio de la concentración y bioacumulación de hidrocarburos policíclicos

aromáticos en doradas procedentes de la acuicultura.


206


207


Como se ha comentado con anterioridad, los PAHs son compuestos orgánicos

ampliamente distribuidos en el medio acuático como resultado de numerosas actividades

humanas y de procesos naturales (Menzie et al., 1992). Estos compuestos orgánicos

también han adquirido una notable importancia como contaminantes debido a su

persistencia en el medio ya que muchos de ellos son potentes tóxicos, mutágenos y

teratógenos para los organismos acuáticos e incluso para el hombre (IARC, 1987). En

consecuencia, los PAHs han sido considerados contaminantes prioritarios por la Agencia

Americana para la Protección del Medio Ambiente (EPA 1987, 1993) y por la USEPA

(Directiva 2000/60/EC).

Los PAHs llegan al medio ambiente marino mediante dos vías: a partir de deposición

atmosférica y procedentes de las aguas que contienen material particulado capaz de

adsorber PAHs (Latimer and Zheng, 2003). Normalmente, en los procesos de combustión,

los PAHs más ligeros con 2-3 anillos aromáticos están en forma gaseosa, mientras que los

que contienen más anillos bencénicos son o bien semivolátiles o están totalmente

adsorbidos a las partículas de hollín. Posteriormente se transfieren desde la atmósfera al

suelo o al agua superficial tanto por deposición como por sedimentación de las partículas.

Dependiendo de la temperatura, se puede producir reemisión desde el suelo. La movilidad

de estos compuestos en el suelo está determinada por el transporte acuático.

Debido a su persistencia, volatilidad y su fuerte absorción, los PAHs tienden a

acumularse en aquellas matrices afines a sus características lipófilas. Su naturaleza lipofílica

hace que sean fácilmente ingeridos y acumulados por organismos acuáticos mediante la

alimentación.

Las materias primas de origen marino son utilizadas frecuentemente en la acuicultura

para aportar ácidos poliinsaturados. Estas matrices son susceptibles de presentar PAHs que

pueden ser acumulados por los peces y transferidos a niveles superiores dentro de la

cadena trófica. Por lo tanto, en la acuicultura marina, la alimentación de peces con materias

primas de origen marino puede suponer la acumulación de PAHs y otros contaminantes

lipófilos en las especies cultivadas.


208

La producción de peces, moluscos y otros invertebrados en la acuicultura ha

aumentado rápidamente durante las últimas décadas como consecuencia del elevado

consumo de estos y la pesca excesiva que sufren las especies salvajes. En los últimos años la

demanda de productos procedentes de la acuicultura ha tenido un crecimiento muy

pronunciado. Las materias primas, sobre todo harinas de pescado y aceites de pescado,

tuvieron alzas significativas en los precios, debido a la mayor demanda mundial. Pero no

hay que olvidar que éstos se producen a partir de productos pesqueros, que son limitados y

que en el futuro se van a ver enfrentados a una mayor escasez.

Es necesario asegurar un futuro sostenible de la acuicultura y actuar para el

desarrollo y la salud del entorno marítimo. Para ello, es imprescindible aumentar el

conocimiento científico de los aspectos eco-toxicológicos de esta actividad. Para evaluar la

toxicidad de los PAHs se utilizan criterios establecidos sobre factores de equivalencia tóxica

(TEFs) (EPA 2000, Nisbet and Lagoy, 1992) y la referencia de toxicidad expresada como

equivalentes del Benzo(a)Pireno (BaPEs). Estos equivalentes tóxicos suponen una

estimación adecuada para conocer la toxicidad de la suma de los PAHs encontrados en las

muestras del estudio.

En el presente estudio se determinó la concentración de PAHs en materias primas,

piensos y filetes de dorada. Todas las muestras provienen de un estudio experimental que

incluye un ciclo completo de engorde de doradas (Sparus aurata L.) en el marco del

Proyecto AQUAMAX (Sustainable Aquafeeds to Maximise the Health Benefits of Farmed

Fish for Consumers. www.aquamaxip.eu). La información generada a partir de estos

estudios es de gran importancia pues permitirá identificar cuáles son las concentraciones

de estos compuestos y si se produce biomagnificación (BMF) en el individuo durante su

crecimiento, hasta llegar al tamaño comercial. También permitirá conocer cuáles son las

concentraciones finales que puede presentar un filete de dorada una vez alcanzado su

tamaño comercial cuando es destinado para el consumo humano.

La determinación de PAHs en las muestras se llevó a cabo mediante metodología

analítica desarrollada como parte de esta Tesis (ver Capítulo 4. Artículo científico 1).

Gracias a esta metodología fue posible eliminar gran parte de los interferentes procedentes


209

de las muestras y determinar PAHs a niveles de µg/kg mediante GC-(QqQ)MS/MS y GC-TOF

MS.


210

Capítulo 4. Bioacumulación de PAHs

211


Bioaccumulation of Polycyclic Aromatic Hydrocarbons in Gilthead Sea Bream (Sparus aurata

L.) Exposed to Long Term Feeding Trials with Different Experimental Diets.

J. Nácher-Mestre, R. Serrano, L. Benedito-Palos, J. C. Navarro, F. J. López, S. Kaushik, J.

Pérez-Sánchez

Archives of Environmental Contamination and Toxicology 59 (2010) 137–146

Archives of Environmental Contamination and Toxicology, 59 (2010) 137–146

212


213

Abstract Polycyclic aromatic hydrocarbons (16 EPA list) were determined in oils, fish

feed, and fillets from gilthead sea bream fed through a full production cycle (14

months) with feed containing different proportions of fish oil replaced by vegetable

oils, followed by a finishing phase with fish oil. At the beginning of the study, fish

presented 46.6 μg/kg fresh weight of the sum of PAHs in fillet and a benzo[a]pyrene

equivalent value of 9.1 μg/kg fresh weight. These levels decreased after 330 days of

rearing to values around 2 μg/kg. Although the concentration increased again during

the finishing phase, they remained low. These low concentrations of PAHs could be the

result of a dilution process associated with fish growth and with the detoxification

pathways, both favored by the low levels of PAHs present in the feeds and the lack of

any other potential source of contamination during the whole rearing period.

Polycyclic aromatic hydrocarbons (PAHs) are compounds formed by two or more fused aromatic

rings originating from natural and anthropogenic sources, such as incomplete combustions,

industrial incinerations, transport, or uncontrolled spills. Sixteen of them have been included in the

list of priority pollutants by the US Environmental Protection Agency (EPA) as a consequence of their

potential adverse effects on organisms, including human health (EPA 1987). Over recent years, a

concern has arisen about the impact of these contaminants in coastal environments and on human


214

consumers, especially as a consequence of accidental spills (Cortazar et al. 2008). They are

ubiquitous in the marine environment, and due to their hydrophobic nature, they tend to be

absorbed rapidly on suspended materials and sediment, becoming bioavailable to fish and other

marine organisms through the food chain (Latimer and Zheng 2003; Liang et al. 2007; Perugini et al.

2007).

Polycyclic aromatic hydrocarbons are carcinogenic in mammalian species and in some fish

species (Hendricks et al. 1985; Schultz and Schultz 1982). Many chemical carcinogens are

procarcinogens requiring biotransformation frequently by oxidative metabolism through the

cytochrome P-450 monooxygenase system (Stegeman and Lech 1991). Metabolization and

depuration ability of PAHs by fish has been investigated and stated by several authors (Ferreira et al.

2006; Kennedy et al. 2004; Martin-Skilton et al. 2008).

Marine aquaculture has seen strong development in the last few decades in order to meet

the increased fish consumption by the world population and the decreasing wild stocks. In

populations having high seafood consumption, both farmed and wild seafood products can also

contribute toward contaminant load of the food basket (Martí-Cid et al. 2007, 2008). Fish used as

raw material for the manufacture of fish oil (FO) and fish meal as feed ingredients are a possible

source of PAHs in fish feed that could possibly be bioaccumulated by farmed fish (Hellou et al. 2005).

Likewise, vegetable oils used in fish feed manufacture are another possible source of PAHs (Moret

and Conte 2000). Aquaculture products are subject to increasing strict control and regulation. The

European Commission Regulation (EC) 1881/2006 of December 2006 has fixed maximum levels of 2

μg/kg fresh weight for benzo[a]pyrene (BaPy) in fish. Additionally, some countries have adopted a

legal limit of 1 μg/kg for BaPy content in smoked fish foodstuff (Bories 1988). However, there appear

to be no legal limits for vegetable oils (Moret and Conte 2000).

In order to estimate the toxicity of PAHs, it is common to apply toxic equivalent factors (TEFs)

for each individual congener in comparison with the carcinogenic activity of BaPy. For comparative

purposes, the PAH concentrations are summed and also expressed as benzo[a]pyrene equivalents

(BaPEs), their relative concentrations being weighted in relation to the carcinogenic potential of

individual PAH compounds using equivalency factors (Law et al. 2002). The BaPEs values calculated

represent the toxicity of the charge load of PAHs in each sample (EPA 2000; Nisbet and LaGoy 1992).

A wide variety of environmental pollutants have been detected in farmed fish (Bordajandi et

al. 2006; Easton et al. 2002; Hites et al. 2004; Liang et al. 2007; Maule et al. 2007; Perugini et al.

2007; Santerre et al. 2000; Serrano et al. 2008a, 2008b). Fish meal and FO commonly used as feed

ingredients, especially from the northern hemisphere, are known to contain persistent organic


215

pollutants (POPs) at different levels and a reduction of these POPs is an issue of concern (Oterhals

and Nygård 2008; Oterhals et al. 2007).

The use of plant-based ingredients as alternatives to fish meal and FO can potentially reduce

the charge load of lipophilic contaminants in aquafeeds, and thereby in farmed fish, improving at the

same time the sustainability of marine fishery resources (Bell et al. 2005; Berntssen et al. 2005;

Bethune et al. 2006). Although the presence and accumulation of organochlorine compounds in

farmed fish have been studied rather extensively (Bordajandi et al. 2006; Easton et al. 2002; Hites et

al. 2004; Jacobs et al. 1998; Santerre et al. 2000; Serrano et al. 2008a, 2008b), very little information

is available on the behavior of PAHs in the aquaculture food chain, apart from guidances on risk

assessment for animal and humans (Commission Regulation 2006; EPA 1993, 2000).

This work is part of a broader study aiming at analyzing the effects of FO substitution by

vegetable oils in low-fish-meal diets. Whereas other parts of the study have dealt with fatty acid

metabolism (Benedito-Palos et al. 2009) and the bioaccumulation of organochlorine compounds

(Nácher-Mestre et al. 2009a), the main objective of the present study was to follow the potential

accumulation of PAHs in fast-growing juveniles of the gilthead sea bream (Sparus aurata L.), a major

farmed fish in the Mediterranean aquaculture, exposed through the entire productive cycle to feeds

with graded levels of FO replaced by a mixture of vegetable oils and then switched for 3 months to a

finishing feed based on FO. The levels of PAHs and their equivalence as BaPEs were determined in

oils, feeds, and fish fillets in order to gather information on the effects of dietary oil sources on end-

product quality and safety.

Materials and Methods

Experimental Diets

Three (isoproteic, isolipidic, and isoenergetic) feeds were formulated with a low inclusion level (20%)

of fish meal and fish soluble protein concentrates. Fish oil from the southern hemisphere was the

only lipid source in the control diet (FO), which was also used during the finishing period. The two

remaining diets contained a blend of vegetable oils (2.5 rapeseed oil: 8.8 linseed oil: 3 palm oil),

replacing 33% (33VO) and 66% (66VO) of the FO. All diets were manufactured using a twin-screw

extruder (Clextral, BC 45) at the Institut National de la Recherche Agronomique (INRA) experimental

research station of Donzacq (Landes, France), dried under hot air, sealed, and kept in air-tight bags

until use. Ingredients and approximate composition of the feeds are reported in Table 1.


216

Bioaccumulation Experiment

Gilthead sea bream obtained from Ferme Marine de Douhet, Ile d’Oléron, France were acclimatized

at the Instituto de Acuicultura de Torre de la Sal (IATS) for 20 days before the start of the study. Fish

of ~18 g initial mean body weights were allocated into 9 fiberglass tanks (3000 L) in groups of 150

fish per tank. Water flow was 20 L/min and the oxygen content of outlet water remained above 85%

saturation. The growth study was undertaken over 14 months (July 11, 2006 to September 2, 2007).

The photoperiod and water temperature (10–26°C) varied over the course of the trial following

natural changes at IATS latitude (40°5′N; 0°10′ E).

During the first 11 months of the trial, the three diets were randomly allocated to triplicate

groups of fish and feed was offered by hand to apparent visual satiety. During the finishing feeding

phase (12 weeks, June 6, 2007 to September 2, 2007), two tanks of the 33VO and two of the 66VO

groups were fed with the FO diet. These groups were then renamed 33VO/FO and 66VO/FO,

respectively. Fish fed the FO diet and one tank each of fish fed 33VO and 66VO diets were

maintained on the initial diets until the end of the study. This led to five treatments in the

bioaccumulation study: FO, 33VO, 66VO, 33VO/FO, and 66VO/FO (Fig. 1).

At the beginning and at regular intervals throughout the finishing feeding phase (0, 330, 360,

390, and 420 days), randomly selected fish (eight per treatment) were killed by a blow on the head

prior to tissue sampling, and the left-side fillet (with skin and bone removed) was excised and stored

at −20°C until analysis. Body weight and fillet yield were not affected by the dietary treatments over

the course of the feeding trial (Benedito-Palos et al. 2009). Seawater for fish culture was checked

using the methodology for PAHs described by Pitarch et al. (2007) and no PAH was detected (limit of

detection: 5 and 100 ng/L), thus ensuring that no known PAH exposure other than that from feeds

was expected during the study.

Analytical Method

Polycyclic aromatic hydrocarbons included in the list of priority contaminants by the US EPA (EPA

1987) were analyzed in oils, fish feed, and fish fillets following the method described by Nácher-

Mestre et al. (2009b). Samples were collected and stored at −20°C until analysis. Eight fish from each


217

Table 1. Ingredients and chemical composition of experimental diets.

FO 33VO 66VO

Ingredient (%)

Fish meal (CP 70%) a 15 15 15

CPSP 90 b 5 5 5




Fish oil c 15.15 10.15 5.15



Palm oil 0 1.25 2.5

Soya lecithin 1 1 1

Binder 1 1 1

Mineral premix d 1 1 1

Vitamin premix e 1 1 1

CaHPO4.2H2O (18%P) 2 2 2

L-Lysine 0.55 0.55 0.55

Composition

Dry matter (DM %) 93.13 92.9 92.77

Protein (% DM) 53.2 52.81 52.62

Fat (% DM) 21.09 21 20.99

Ash (% DM) 6.52 6.69 6.57

aFish meal (Scandinavian LT) bFish soluble protein concentrate (Sopropêche, France) cFish oil (Sopropêche, France) dSupplied the following (mg · kg /diet, except as noted): calcium carbonate (40% Ca) 2.15 g, magnesium hydroxide (60% Mg) 1.24 g, potassium chloride 0.9 g, ferric citrate 0.2 g, potassium iodine 4 mg, sodium chloride 0.4 g, calcium hydrogen phosphate 50 g, copper sulphate 0.3, zinc sulphate 40, cobalt sulphate 2, manganese sulphate 30, sodium selenite 0.3. eSupplied the following (mg · kg /diet): retinyl acetate 2.58; DL-cholecalciferol, 0.037; DL-α tocopheryl acetate, 30; menadione sodium bisulphite, 2.5; thiamin, 7.5; riboflavin, 15; pyridoxine, 7.5; nicotinic acid, 87.5; folic acid, 2.5; calcium pantothenate, 2.5; vitamin B12, 0.025; ascorbic acid, 250; inositol, 500 ; biotin, 1.25; choline, chloride 500


218

July 2006 Sept 2007June 2007

Fig. 1 Schematic representation of the long-term feeding trial over 14 months

treatment were randomly selected to obtain three composite samples of three, three, and two

fillets. Fish diets and oils were homogenized and analyzed in triplicate.

Before analysis, samples were thawed at room temperature and were carefully ground (fillets

and feeds) using a kitchen grinder (Super JS, Moulinex, France). Approximately 2 g of sample were

homogenized with 6 g of anhydrous sodium sulfate and the blend was spiked with 100 μL of

surrogate solution (25 ng/mL) Ten milliliters of methanolic solution of 1 M KOH was added to the

mixture and saponified for 3 h at 80°C. Then analytes were extracted twice with 8 mL of n-hexane

and the solution filtered through 0.2 μm and concentrated under gentle nitrogen stream at 40°C to 1

mL (in the case of oils, extracts were concentrated to 5 mL). One milliliter of the extract was passed

through the Florisil SPE cartridge previously conditioned with 6 mL of n-hexane and eluted with 8 mL

dichloromethane:n-hexane (20:80) solution. The eluate was evaporated and redissolved in 0.25 mL

of n-hexane. The final extracts obtained after the cleanup procedure were analyzed by means of a

triple quadrupole analyzer [details in Nácher-Mestre et al. (2009b)]. Quantification of samples was

carried out by means of external calibration curves using the internal standard method. Statistical

validation of the method developed was performed by evaluating analytical parameters. The method

was applied to the four matrixes studied at different fortification levels offering satisfactory


219

recoveries (between 70% and 120%) and precisions (<30%) in all cases [more details in Nácher-

Mestre et al. (2009b)].

The whole analytical process was carried out in Good Laboratory Practices-certified

laboratories of the Research Institute for Pesticides and Water, University Jaume I, Castellón, Spain.

Determination of Fat

The total fat content in the sample extracts was determined gravimetrically, after evaporation at

95°C until constant weight.

Data Analysis

The PAH concentrations are expressed as micrograms per kilogram of fresh weight and as

micrograms per kilogram of lipid base in fillets. Bioaccumulation factors (BAFs) were calculated as

the ratio between lipid-based concentrations of PAHs both in the fish fillets and in the fish feed at

each sampling point. Because two different fish feed were used in the 33VO/FO and 66VO/FO

groups, calculations of BAFs were made correcting for respective feed intakes. Thus, the arithmetic

mean of PAH concentrations in the diet was calculated on the basis of the gram of dry matter

ingested per fish, and the concentration of PAHs in each fish feed. BaPEs were calculated for fish

fillets using the TEFs proposed by the US EPA (EPA 2000). The concentrations of individual PAHs

(μg/kg fresh weight) are expressed as equivalents of BaPy, and these values are summed to obtain a

BaPEs value. Data for concentrations of PAHs were compared by means of ANOVA I and a posteriori

Scheffe’s test (p < 0.05). All data were log-transformed before statistical analysis to achieve

normality. Homoscedasticity of variances was tested by means of Bartlett’s test (p < 0.05). All of the

statistical analyses were conducted using STATGRAPHICS plus for Windows version 4.1 (Statistical

Graphics Corporation).

Results and Discussion

PAHs in Oils and Feeds

The analytical method showed excellent sensitivity and selectivity as a consequence of the use of gas

chromatography-coupled tandem mass spectrometry (GC–MS/MS) together with the efficiency of


220

the saponification followed by solid-phase extraction clean up. It has allowed analyte quantification

at concentrations as low as picograms per gram of fresh weight and micrograms per kilogram of lipid

weight in the case of oils, with acceptable precisions (below 30%) (Nácher-Mestre et al. 2009b).

Concentrations of PAHs present in the oils used in the feeds are shown in Table 2. Up to 14 of

the 16 PAHs considered were detected only at trace levels. Both FO and vegetable oils had similar

levels and patterns, with individual PAHs concentrations up to 38.2 μg/kg fresh weight and total

loads (Σ16PAH) of 56.7 μg/kg fresh weight in FO and between 12 and 47 μg/kg fresh weights in

vegetable oils. The presence of PAHs in the marine environment has been reported by several

authors (Liang et al. 2007; Oros and Ross 2005; Perugini et al. 2007). Vegetable derivatives can

accumulate PAHs from different sources such as atmospheric deposition of contaminated dust and

particulate matter in the plants and the processing for oil production (Moret and Conte 2000).

The PAH concentrations in the feeds did not correlate with the amounts present in the FO or

vegetable oils (VOs) used. We find here that replacement of FO by VOs in the feed did not allow one

to minimize the PAHs charge load in complete fish feed. It then appears that the PAHs found in the

feed originated from ingredients other than the oils used. The contamination pattern was similar in

the different feeds with a predominance of 2-ring PAH naphthalene (up to 215 μg/kg fresh weight)

and 3- and 4-ring PAHs, whereas 5- and 6-ring PAHs were less abundant. Naphthalene is the PAH

with the lowest K ow and bioconcentration factor (BCF) values. Liang et al. (2007) have reported a

similar PAH pattern contamination in fish, with trace concentrations of individual PAHs but

naphthalene at concentrations higher than 100 μg/kg fresh weight. In that particular case, this was

related to the high accumulation factor biota-sediment calculated for this compound. In our case,

the sources and behavior of these compounds are very complex. The different ingredients used in

the feeds are processed products from fishery, plant, and industrial sources and the proportions of

the different individual PAHs, which could change during the manufacturing process. Even the drying

of raw materials by means of hot air could be a potential source of contamination (Moret and Conte

2000). This process is always followed by a considerable increase in PAH content. Moret et al. (2000)

reported that oils could have dangerous levels of PAHs for human health due to the drying process.

Specifically, BaPy presented an increased concentration in oil samples after drying, ranging from 8.6

to 44.3 ppb (Moret et al. 2000).


221

Table 2 Concentration of polycyclic aromatic hydrocarbons (PAHs, µg/kg) in the oils used and in the complete feeds

Ingredients Fish oil Linseed oil Rapeseed oil Palm oil FO 33VO 66VO

Compound Mean CV Mean CV Mean CV Mean CV Mean C.V. Mean C.V. Mean C.V.

Naphthalene - - 1.9 22 0.8 24 215 11 242 10 161 9

Acenaphthylene - - - - 0.6 16 0.6 4 0.8 18

Acenaphthene 0.3 8 - 0.6 13 0.5 15 1.7 7 1.4 6 1.2 8

Fluorene 4.3 5 2.6 4 0.9 4 0.4 8 3.8 4 4 10 3.5 8

Phenanthrene 38.2 2 15.2 1 1.2 3 1.3 3 6.1 3 5.6 1 5 5

Anthracene - - 1.6 7 - 1.1 21 1 12 1.3 3

Fluoranthene 13.3 7 16.7 4 - - 2.1 3 2 7 2 4

Pyrene 9 1 11.3 1 - - 2.1 15 1.9 11 1.8 10

Benzo(a)Anthracene - - 0.9 1 1.2 1 - - -

Chrysene - - 0.8 4 - 0.6 6 0.8 22 0.6 15

Benzo(b)Fluoranthene 0.3 22 0.5 4 0.4 4 1.3 3 0.7 13 0.6 20 0.7 16

Benzo(k)Fluoranthene 0.3 2 0.4 5 0.4 9 1.3 7 0.6 25 0.6 9 0.5 5

Benzo(a)Pyrene - 0.3 8 0.7 4 1.4 3 - - -

Indeno(1,2,3-cd)Pyrene - - 0.7 7 1.1 6 - - -

Dibenzo(a.h)Anthracene - - 0.9 2 1.3 1 - - -

Benzo(g.h.i)Perylene - - 0.8 8 1.2 6 - - -

ΣΣΣΣPAH16 56.7 31 47 31 12 2 12 2 234.4 44.2 260.5 39.1 178.4 34.5

CV: coefficient of variation. Compounds were quantified with CV<30%. -: non detected. FO: Fish oil diet. 33VO: 33 % vegetable oil diet. 66VO: 66%

vegetable oil diet.


222

PAHs in Fish Fed the Different Feeds

Table 3 presents the individual concentrations and Σ16PAHs in fillets from fish fed the different

experimental diets during the study. Initial fish had 46.6 μg/kg fresh weight of Σ16PAHs (1084 μg/kg

lipid). After 330 days of feeding with the diets, values descended to 1.1–1.9 μg/kg fresh weight

(14.6–24.4 μg/kg lipid) in the different groups. These low concentrations of PAHs remained at trace

levels until the end of the study. Naphthalene was not detected after 11 months of rearing, probably

as consequence of its physicochemical characteristics, which could explain the low bioaccumulation

ability found here. Similarly, anthracene, benzo[a]anthracene and all 5- and 6-ring PAHs initially

present in fish at time 0 were not detected after 330 days of exposure.

Data on BaPEs in the fillets from fish fed the different feeds are also reported in Table 3. BaPE

values were below 0.02 μg/kg fresh weight, except for initial fish, which had a value of 9.1 μg/kg

fresh weight, which is higher than the maximum level of 2 μg/kg allowed in fish by the Commission

Regulation (EC) 1881/2006 (Commission Regulation 2006).

The decrease of PAH concentrations (expressed per unit fresh weight or unit lipid) during the

first 11 months suggests the action of a dilution process linked to body mass increase (from 18 to

300 g) in the absence of PAH sources other than the low load from diets. This process does not

preclude a contaminant biodepuration of PAHs present in initial fish through detoxification

metabolic routes. The load of organochlorine pollutants during the same growth study was also

monitored (Nácher-Mestre et al. 2009a) and did not reveal any accumulation during the first 11

months of feeding with the different feeds.

As reported in Fig. 2, in spite of the low levels of PAHs found in fish fillets during the study,

the ∑16PAHs concentration appreciably increased after 360 days of exposure in all groups, coinciding

with the increase in feed intake and the growth spurt of summer (from 290 to 530 g in 90 days).

These differences were significant at 390 days (Scheffe’s test, p < 0.05), especially in groups refed

with diet FO and were followed by a plateau by the end of the finishing phase.

The load of POPs, including organochlorine pesticides, polychlorinated byphenyls, PAHs, and

polybrominated diphenyl ethers, has also been characterized in the feeds as part of the project. The

total load of contaminants (POPs) in diet FO (Nácher-Mestre et al. 2009a) was higher than that of the

other feeds. This could lead to a decrease in the efficacy of the PAH metabolism in fish from groups

refed FO the last 3 months. On the other hand, the stabilization of the concentrations during the last

month of the trial suggests the adaptation of the fish to the uptake levels of pollutants presented the

last 3 months of the experiment and the recovery of the ability to biodepurate the PAHs up taken via


223

Table 3 Concentrations of PAHs (μg/kg fresh weight; n = 3) in fillets from gilthead sea bream fed the different diets

Diet FO 33VOFO 66VO 66VOFO Time of exposure T0 T1 T4 T1 T4 T4 T1 T4 T4

Compound

Mean CV Mean CV Mean CV Mean CV Mean CV Mean CV Mean CV Mean CV Mean CV

Naphthalene 3.9 1 - - - - - - - - Acenaphthylene - - - - - - - - - Acenaphthene - - - - - - - - - Fluorene - 0.36 2 0.51 8 0.32 17 0.63 29 0.52 28 0.22 14 0.58 6 0.41 5 Phenanthrene 2.0 1 0.86 26 1.83 7 0.79 29 2.11 27 2.40 10 0.49 7 2.21 18 2.13 17

Anthracene 1.0 1 - - - - - - - - Fluoranthene 3.9 1 0.28 12 0.99 13 0.22 16 1 7 1.8 22 0.18 3 0.64 29 1.61 11 Pyrene 4.9 1 0.4 27 2.8 19 0.31 19 2.07 1 4.71 27 0.25 9 1.35 29 4.21 12 Benzo(a)Anthracene 4.9 1 - - - 0.09 13 - - 0.14 2 0.15 13

Chrysene 6.9 1 - 0.14 13 - 0.17 14 0.16 8 - 0.16 4 0.18 20 Benzo(b)Fluoranthene 5.9 1 - - - - - - - - Benzo(k)Fluoranthene 5.9 1 - - - - - - - - Benzo(a)Pyrene 3.9 1 - - - - - - - -

Indeno(1,2,3-cd)Pyrene - - - - - - - - - Dibenzo(a.h)Anthracene - - - - - - - - - Benzo(g.h.i)Perylene 3.4 1 - - - - - - - -

ΣΣΣΣPAH16 (µg/kg resh 46.6 3 1.9 20 6.3 7 1.6 19 6.06 7 9.6 26 1.1 12 5.1 14 8.7 6

ΣΣΣΣPAH16 (µg/kg lipid 1084 36 24.4 9 68.2 10 21.6 9 60.6 9 85.8 7 14.6 3 50.8 9 79.7 9 BaPEs (µg/kg fresh 9.1 2 0.002 10 0.007 15 0.002 13 0.002 19 0.013 16 0.001 6 0.006 15 0.010 21

T0 : July/06. T1 : June/07. T4 : Sept/07. Mean: arithmetic mean. CV: coefficient of variation (%). For the rest of abbreviations, see Figure 1.


224

Fig. 2 Concentration of ∑16PAH in fillets of fish fed different experimental diets with varying levels of fish oil substitution, during the feeding trial. Letters denote significant differences among sampling times for all groups; Scheffe’s test; p < 0.05

diet. The extent to which this biodepuration process coexists with the above-mentioned dilution

mechanism and the relative efficiency of both, especially foreseen in shorter periods of higher

toxicant intake, like the finishing phase, remains to be ascertained.

Although it should be kept in mind that the concentrations present in fish are at the trace

level, the increase of the PAH content at T4 could be explained in light of the increase of feed intake

(Fig. 1) during the finishing phase and the reaching of a PAH intake level higher than the

detoxification route capability.

In comparison with concentrations reported in the literature regarding seafood samples

(Fontcuberta et al. 2006; Llobet et al. 2006; Saeed et al. 1995), the ∑PAHs values determined in

gilthead sea bream as found here after 330 days of culture were low (~2 μg/kg). Although the

concentration increased again during the finishing phase, as we have indicated earlier, it remained

low, below 10 μg/kg in all groups, and similar to those found in fish from unpolluted areas. DouAbdul

et al. (1997) indicated concentrations from <0.5 to 148 μg/kg fresh weight in fish fillets from

0

2

4

6

8

10

12

330 340 350 360 370 380 390 400 410 420time of exposure (days)

FO 33VO 66VO 33VOFO 66VOFO

A A

B

B

B

B

0

2

4

6

8

10

12



A A

B

B

B

B

µg/k

g fr

esh

wei

ght S

um P

AH

16

0

2

4

6

8

10

12



A A

B

B

B

B

0

2

4

6

8

10

12



A A

B

B

B

B

0

2

4

6

8

10

12



A A

B

B

B

B

0

2

4

6

8

10

12



A A

B

B

B

B

µg/k

g fr

esh

wei

ght S

um P

AH

16


225

unpolluted areas. Bordajandi et al. (2004) reported levels between 8.2 and 71.4 μg/kg fresh weight

in marine food samples from Spain, whereas Loufty et al. (2007) found levels in market fish samples

from Egypt ranging from 0.78 to 19.70, similar to those observed in gilthead sea bream in this study

and very low in comparison to polluted areas.

As reported in Table 4, individual PAHs found in fish fillets show variable BAF values.

Phenanthrene, fluoranthene, and pyrene showed BAFs over 1 in almost all groups of fish.

Phenanthrene, fluoranthene, and pyrene are bioaccumulated probably as a consequence of their

high K ow (Table 4) because possible sources of these compounds other than diet, like combustions

or petroleum, can be discarded. Nevertheless, chrysene had BAF values below 1 in spite of its higher

K ow, which could be interpreted in terms of a more efficient degradation or depuration of this

compound. On the basis of data obtained here, the different BAF values found for the different

individual PAHs can be attributed to the variety of structures and physicochemical properties of the

different components of the family.

Table 4 Bioaccumulation factors (BAF) for the polycyclic aromatic hydrocarbons (PAHs) detected after 420 days of the growth trial with the different diets (from lipid based concentrations in fish fillets and diets, n=3)

Diet

FO 33VO

33VOFO 66VO 66VOFO Kow

Compound BAF CV BAF CV BAF CV BAF CV BAF CV

Fluorene 0.29 8 0.32 29 1.10 6 0.26 28 0.98 5 4.18

Phenanthrene 0.65 7 0.75 27 2.45 18 1.21 10 1.63 17 4.57

Fluoranthene 1.02 13 1.00 7 2.28 29 1.80 22 2.42 11 5.22

Pyrene 2.90 19 2.18 1 4.05 29 6.01 27 3.60 12 5.18

Chrysene 0.51 13 0.43 14 0.52 8 0.49 4 0.91 20 5.86

Sources

The study of three PAH isomer ratios—anthracene/anthracene + phenanthrene (An/178),

benz[a]anthracene/benz[a]anthracene + chrysene (BaA/228), and fluoranthene/fluoranthene +


226

pyrene [Fl/(Fl + Py)—was applied to identify the possible major sources of these contaminants in the

different matrixes considered in this work (Table 5). These PAH ratios can be used as indicators of

the different formation processes of the PAH contamination and of the possible sources of

contamination in sediments (Yunker et al. 2002). FO and VOs have An/178 ratios below 0.1, which

suggest unburned petroleum as one of the sources of the contamination present in these products.

These An/178 ratios determined in the oils are similar to those found by Yunker et al. (2002) in

sediments from remote or light-urban areas. The FO used in this research was from South American

origin, which could explain the presence of PAHs from unburned petroleum. Fl/(Fl + Py) and BaA/228

ratios indicate nonpetroleum combustion sources of the PAHs found in the samples, supporting the

idea of PAH contamination from light-urbanized areas.

Table 5 PAH isomer ratios (mean ± SD, n = 6) in oils, diets, and fish fed the different diets

a An/178: anthracene/anthracene+phenanthrene,

b

BaA/228: benzo[a]anthracene/benzo[a]anthracene+chrysene c FL/FL + Py: fluoranthene/fluoranthene+pyrene,

dPhe/An: phenanthrene/anthracene,

e

Flu/Pyr: fluoranthene/pyrene f data for fish fillets are the mean of all experimental groups

The feeds manufactured with the above-studied ingredients showed higher An/178 values,

probably as a consequence of the manufacture process itself that could provoke degradations and

PAH isomer ratio An/178a BaA/228

b FL/FL+Py

c Phe/An

d Flu/Pyr

e

Product

Vegetable oils 0.01±0.01 0.56±0.06 0.60±0.05 76 ± 10 1.5 ± 0.2

FO diet 0.15±0.02 0 0.50±0.06 5.5 ± 1.1 1.0 ± 0.2

33VO diet 0.15±0.01 0 0.51±0.05 5.6 ± 0.7 1.0 ± 0.1

66VO diet 0.21±0.02 0 0.53±0.06 3.8 ± 0.2 1.1 ± 0.1

Fish fillets T0f 0.33±0.02 0.42±0.05 0.44±0.06 2.0 ± 0.02 0.8 ± 0.008

Fish fillets T1 0 0 0.41±0.07 0 0.7 ± 0.2

Fish fillets T2 0 0 0.43±0.06 0 0.8 ± 0.01

Fish fillets T3 0 0 0.35±0.05 0 0.2 ± 0.001

Fish fillets T4 0 0 0.26±0.07 0 0.3 ± 0.06


227

contaminations of the different individual PAHs. However, it is interesting to note that

benzo[a]anthracene was not detected in the feed, possibly due to the degradation process during

manufacture, as this PAH is one of the less stable (Yunker et al. 2002). The Fl/(Fl + Py) ratios found in

the feeds are similar to those of the oils used: below 0.5.

Initial fish showed values for An/178, BaA/228, and Fl/(Fl + Py) of 0.333, 0.415, 0443,

respectively. These values indicate both petroleum combustion and other combustions as sources of

the PAHs in sediments (Yunker et al. 2002) and bivalves (Oros and Ross 2005), accumulated prior to

the beginning of the present study. The An/178 and BaA/228 ratios decreased to 0 after 11 months

of culture, whereas Fl/(Fl + Py) values decreased continuously during the whole growth period,

suggesting a decrease of PAH exposure during the experimental period.

Conclusions

The levels of PAHs found in fish fillets grown with different diets over a full production cycle were

very low and similar to those reported in foods from unpolluted marine areas. These low levels

correspond to low levels of PAHs found in the diets and the lack of other potential sources of

contamination in the rearing facilities over the experimental period. These conditions together with

dilution and biodepuration mechanisms result in a net decrease of total PAH loads present in

experimental fish at the start of the experiment and, thus, in the production of market size fish of

good quality for human consumption.

Acknowledgment This study was carried out with the financial support of the European Union

Project: Sustainable Aquafeeds to Maximise the Health Benefits of Farmed Fish for Consumers

(AQUAMAX), contract number 016249-2.

References

Bell JG, McGhee F, Dick JR, Tocher DR (2005) Dioxin and dioxin-like polychlorinated biphenyls (PCBs)

in Scottish farmed salmon (Salmo salar): effects of replacement of dietary marine fish oil with

vegetable oils. Aquaculture 243:305–314

Benedito-Palos L, Navarro JC, Bermejo-Nogales A, Saera-Vila A, Kaushik S, Pérez-Sánchez J (2009) The

time course of fish oil wash-out follows a simple dilution model in gilthead sea bream (Sparus

aurata L.) fed graded levels of vegetable oils. Aquaculture 288:98–105


228

Berntssen MHG, Lundebye AK, Torstennsen BE (2005) Reducing the levels of dioxins and dioxin-like

PCBs in farmed Atlantic salmon by substitution of fish oil with vegetable oil in the feed.

Aquacult Nutr 11:219–231

Bethune C, Seierstad SL, Seljeflot I, Johansen O, Arnesen H, Meltzer HM, Rosenlund G, Froyland L,

Lundebye AK (2006) Dietary intake of differentially fed salmon: a preliminary study on

contaminants. Eur J Clin Invest 36:193–201

Bordajandi LR, Gómez G, Abad E, Rivera J, Del Mar Fernandez-Baston M, Blasco J, González MJ (2004)

Survey of persistent organochlorine contaminants (PCBs, PCDD/Fs, and PAHs), heavy metals

(Cu, Cd, Zn, Pb, and Hg), and arsenic in food samples from Huelva (Spain): levels and health

implications. J Agric Food Chem 52:992–1001

Bordajandi LR, Martín I, Abad E, Rivera J, Gonzalez MJ (2006) Organochlorine compounds (PCBs,

PCDDs and PCDFs) in sea fish and seafood from the Spanish Atlantic Southwest Coast.

Chemosphere 64:1450–1457

Bories G (ed) (1988) Tossicologia e sicurezza degli alimenti. Tecniche Nuove, Milan, p 359

Commission Regulation (2006) Commission Regulation (EC) No 1881/2006 of 19 December 2006

setting maximum levels for certain contaminants in foodstuffs. Available from

(eurlex.europa.eu/LexUriServ/site/en/oj/2006/l_364/l_36420061220en00050024.pdf)

Cortazar E, Bartolomé L, Arrasate S, Usobiaga A, Raposo JC, Zuloaga O, Etxebarria N (2008)

Distribution and bioaccumulation of PAHs in the UNESCO protected natural reserve of Urdaibai,

Bay of Biscay. Chemosphere 72:1467–1474

Directiva 2000/60/CE del Parlamento Europeo y del Consejo, de 23 de octubre de 2000, por la que se

establece un marco comunitario de actuación en el ámbito de la política de aguas.

DouAbdul AAZ, Heba HMA, Farred KH (1997) Polynuclear aromatic hydrocarbons (PAHs) in fish from

the Red Sea Coast of Yemen. Hydrobiologia 352:251–262

EPA (1987) Quality criteria for water. EPA 440/5–86-001. US Environmental Protection Agency,

Washington DC

EPA (1993) Provisional guidance for quantitative risk assessment of polycyclic aromatic

hydrocarbons. EPA 600/R-93/089. US Environmental Protection Agency, Washington DC

EPA (2000) Guidance for assessing chemical contaminant data for use in fish advisories. EPA. 823-B-

00–008. US Environmental Protection Agency, Washington DC

Easton MDL, Luszniak D, Von der Geest E (2002) Preliminary examination of contaminant loadings in

farmed salmon, wild salmon and commercial salmon feed. Chemosphere 46:1053–1074


229

Ferreira M, Moradas-Ferreira P, Reis-Henriques MA (2006) The effect of long-term depuration on

phase I and phase II biotransformation in mullets (Mugil cephalus) chronically exposed to

pollutants in River Douro Estuary, Portugal. Marine Environ Res 61:326–338

Fontcuberta M, Arqués JF, Martínez M, Suárez A, Villalbí JR, Centrich F, Serrahima E, Duran J, Casas C

(2006) Polycyclic aromatic hydrocarbons in food samples collected in Barcelona, Spain. J Food

Prot 69:2024–2028

Hellou J, Haya K, Steller S, Burridge L (2005) Presence and distribution of PAHs, PCBs and DDE in feed

and sediments under salmon aquaculture cages in the Bay of Fundy, New Brunswick, Canada.

Aquat Conserv 15:349–365

Hendricks JD, Meyeres TR, Shelton DW, Casteel JL, Bailey GS (1985) Hepatocarcinogenicity of

benzo(a)pyrene to rainbow trout by dietary exposure and intraperitoneal injection. J Natl

Cancer Inst 74:839–851

Hites RA, Foran JA, Carpenter DO, Hamilton MC, Knuth BA, Schwager SJ (2004) Assessment of organic

contaminants in farmed salmon. Science 303:226–229

IARC, (1987). IARC monographs on the evaluation of carcinogenic risks to humans, supplement 7.

Overall evaluations of carcinogenicity: an updating of IARC monographs, vol 1-42., Vol.

International Agency for Research on Cancer, Lyon.

Jacobs MN, Santillo D, Johnston PA, Wyatt CL, French MC (1998) Organochlorine residues in fish oil

dietary supplements: comparison with industrial grade oils. Chemosphere 37:1709–1721

Kennedy CJ, Higgs D, Tierney K (2004) influence of diet and ration level on benzo[a]pyrene

metabolism and excretion in rainbow trout (Oncorhyncus mykiss). Arch Environ Contam Toxicol

47:379–386

Latimer JS, Zheng J (2003) Sources, transport and fate of PAHs in the marine environment. In:

Douben PET (ed) PAH: an ecotoxicological perspective. Wiley, Chichester, UK, pp 9–33

Law RJ, Kelly C, Baker K, Jones J, McIntosh AD, Moffat CF (2002) Toxic equivalency factors for PAH

and their applicability in shellfish pollution monitoring studies. J Environ Monit 4:383–388

Liang Y, Tse MF, Young L, Wong MH (2007) Distribution patterns of polycyclic aromatic hydrocarbons

(PAHs) in the sediments and fish at Mai Po Marshes Nature Reserve, Hong Kong. Water Res

41:1303–1311

Llobet JM, Falcó G, Bocio A, Domingo JL (2006) Exposure to polycyclic aromatic hydrocarbons

through consumption of edible marine species in Catalonia, Spain. J Food Prot 69:2493–2499


230

Loufty N, Fuerhacker M, Tundo P, Raccanelli S, Ahmed MT (2007) Monitoring of polychlorinated

dibenzo-p-dioxins and dibenzofurans, dioxin-like PCBs and polycyclic aromatic hydrocarbons in

food and feed samples from Ismailia city, Egypt. Chemosphere 66:1962–1970

Martí-Cid R, Bocio A, Llobet JM, Domingo JL (2007) Intake of chemical contaminants through fish and

seafood consumption by children of Catalonia, Spain: health risks. Food Chem Toxicol 45:1968–

1974

Martí-Cid R, Llobet JM, Castell V, Domingo JL (2008) Evolution of the dietary exposure to polycyclic

aromatic hydrocarbons in Catalonia, Spain. Food Chem Toxicol 46:3163–3171

Martin-Skilton R, Saborido-Rey F, Porte C (2008) Endocrine alteration and other biochemical

responses in juvenile trubot exposed to the Prestige fuel oil. Sci Total Environ 404:68–76

Maule AG, Gannam AL, Davis JW (2007) Chemical contaminants in fish feeds used in federal

salmonid hatcheries in the USA. Chemosphere 67:1308–1315

Moret S, Conte LS (2000) Polycyclic aromatic hydrocarbons in edible fats and oils: occurrence and

analytical methods. J Chromatogr A 882:245–253

Moret S, Dudine A, Conte LS (2000) Processing effects on the polyaromatic hydrocarbon content of

grapeseed oil. J Am Oil Chem Soc 77:1289–1292

Nácher-Mestre J, Serrano R, Benedito-Palos L, Navarro JC, Lopez FJ, Pérez-Sanchez J (2009a) Effects

of fish oil replacement and re-feeding on the bioaccumulation of organochlorine compounds in

gilthead sea bream (Sparus aurata L.) of market size. Chemosphere 76:811–817

Nácher-Mestre J, Serrano R, Portolés T, Hernández F, Benedito-Palos L, Pérez-Sanchez J (2009b) A

reliable analytical approach based on gas chromatography coupled to triple quadrupole and

time of flight analyzers for the determination and confirmation of polycyclic aromatic

hydrocarbons in complex matrices from aquaculture activities. Rapid Commun Mass Spectrom

23:2075–2086

Nisbet IC, LaGoy PK (1992) Toxic equivalency factors (TEFs) for polycyclic aromatic hydrocarbons

(PAHs). Regul Toxicol Pharm 16:290–300

Oros DR, Ross JRM (2005) Polycyclic aromatic hydrocarbons in bivalves from the San Francisco

estuary: spatial distributions, temporal trends, and sources (1993–2001). Marine Environ Res

60:466–488

Oterhals Å, Nygård E (2008) Reduction of persistent organic pollutants in fishmeal: a feasibility study.

J Agric Food Chem 56:2012–2020


231

Oterhals Å, Solvang M, Nortvedt R, Berntssen MHG (2007) Optimization of activated carbon-based

decontamination of fish oil by response surface methodology. Eur J Lipid Sci Technol 109:691–

705

Perugini M, Visciano P, Giammarino A, Manera M, Di Nardo W, Amorena M (2007) Polycyclic

aromatic hydrocarbons in marine organisms from the Adriatic Sea, Italy. Chemosphere

66:1904–1910

Pitarch E, Medina C, Portolés T, Lopéz FJ, Hernández F (2007) Determination of priority organic

micro-pollutants in water by gas chromatography coupled to triple quadrupole mass

spectrometry. Anal Chim Acta 583:246–258

Saeed T, Al-Yakoob S, Al-Hashash H, Al-Bahloul M (1995) Preliminary exposure assessment for

Kuwaiti consumers to polycyclic aromatic hydrocarbons in seafood. Environ Int 21:255–263

Santerre CR, Ingram R, Lewis GW, Davis JT, Lane LG, Grodner RM, Wei CI, Bush PB, Xu DH, Shelton J,

Alley EG, Hinshaw JM (2000) Organochlorines, organophosphates, and pyrethroids in channel

catfish, rainbow trout, and red swamp crayfish from aquaculture facilities. J Food Sci 65:231–

235

Schultz M, Schultz J (1982) Induction of hepatic tumors with 7, 12-dimethyl-benzantracene in two

species of viviparous fishes (genus Poecillopsis). Environ Res 27:337–351

Serrano R, Barreda M, Blanes MA (2008a) Investigating the presence of organochlorine pesticides

and polychlorinated biphenyls in wild and farmed gilthead sea bream (Sparus aurata) from the

Western Mediterranean Sea. Marine Pollut Bull 56:963–972

Serrano R, Blanes MA, López FJ (2008b) Biomagnification of organochlorine pollutants in farmed and

wild gilthead sea bream (Sparus aurata) and stable isotope characterization of the trophic

chains. Sci Total Environ 389:340–349

Stegeman JJ, Lech JJ (1991) Cytochrome P-450 monooxygenase systems in aquatic species:

carcinogen metabolism and biomarkers for carcinogen and pollutant exposure. Environ Health

Perspect 90:101–109

Yunker MB, Macdonald RW, Vingarzan R, Mitchell RH, Goyette D, Sylvestre S (2002) PAHs in the

Fraser river basin: a critical appraisal of PAH ratios as indicators of PAH source and composition.

Org Geochem 33:489–515


232


233


La selección de los PAHs objeto de estudio en las muestras analizadas se basó en la

lista propuesta por la USEPA sobre toxicidad (EPA. 1993; EPA. 2000; ATSDR 1995b). El

estudio de los PAHs propuestos nos permite comparar nuestros resultados con trabajos

sobre la acuicultura atlántica en salmones. Esta lista de PAHs incluye compuestos con 2, 3,

4, 5 y 6 anillos de benceno con capacidad tóxica. También está incluido en la lista el

benzo(a)pireno que está considerado como referencia de toxicidad según la documentación

europea (Reglamento 1881/2006/CE). Además, incluye el naftaleno, un PAH muy ubicuo

que presenta grandes dificultades analíticas.

En general, las muestras analizadas presentaron numerosos positivos de PAHs (Tabla

2, 3. Artículo de Investigación 5.). Los PAHs más frecuentemente identificados en las

muestras fueron el fluoreno, fenantreno, fluoranteno y pireno. También son estos los que

presentaron mayor concentración. Hay una excepción con el naftaleno, pues presenta

concentraciones muy altas en piensos en comparación con los demás PAHs cuantificados. El

aceite más contaminado fue, como era de esperar, el aceite de pescado (ΣPAHs= 56.7 µg/Kg).

También presentó concentraciones mayor de lo esperadas el aceite de lino (ΣPAHs= 47

µg/Kg) mientras que el aceite de colza y el aceite de palma presentaron ΣPAHs= 12 µg/Kg en

ambos casos. La mayor concentración en piensos (FO, 33VO, 66VO) se debió

principalmente a la concentración individual del naftaleno, cercana a los 200 µg/Kg en

todos los piensos. Sin tener en cuenta el naftaleno, otros trabajos publicados

recientemente presentan concentraciones de PAHs mucho mayores en comparación a las

encontradas en nuestro trabajo en piensos utilizados en la acuicultura (Tsapakis et al.,

2010).

Los PAHs detectados en los piensos no siguen el mismo perfil de las materias primas

analizadas, sino que presentan un perfil propio y homogéneo entre los diferentes piensos

ensayados. Este hecho puede ser una consecuencia del proceso de manufacturación del

pienso, durante el cual, los ingredientes son sometidos a secado mediante aire caliente

(Moret et al., 2000). Este proceso puede, por una parte degradar los PAHs con más anillos

aromáticos, y por otra parte formar otros PAHs con menor número de anillos como el


234

naftaleno. Como se ha comentado anteriormente, este último está presente en los piensos

a una concentración mucho más elevada que en el resto de muestras analizadas (hasta 215

µg/Kg, peso fresco). Este nivel elevado de naftaleno se apoya también en resultados de

otros investigadores que lo consideran como un PAHs muy ubicuo (Liang et al., 2007). Se

sospecha que el origen de la alta concentración de naftaleno detectada es la quema de

hojarasca que periódicamente tiene lugar en las inmediaciones del Instituto (IATS) donde se

realizaron los experimentos de engorde.

Respecto a los PAHs detectados en los filetes de los peces cultivados, en la Figura 6.1.

se puede observar que la carga total de PAHs decreció considerablemente en los primeros

11 meses de cultivo, a partir de los cuales se produjo una ligera tendencia al alza en todos

los grupos sin alcanzar, en ningún caso, concentraciones mayores a 11 µg/Kg.

inicial

FO

33VO

66VO

33VO FO66VO FO

0

10

20

30

40

50

con

cen

tra

ció

n e

n p

eso

fre

sco

(µg

/Kg

)

ΣPAHs en filetes

Figura 6.1. Niveles de contaminación de los filetes de dorada expresados como ΣPAHs

durante el proceso experimental mediante la ingesta de diferentes piensos experimentales.

Los filetes de dorada no presentaron concentraciones de contaminantes significativas

durante los 11 primeros meses del ciclo de crecimiento. Este hecho puede deberse a los


235

bajos niveles de concentración presentes en las dietas utilizadas. Después de este período

de alimentación las concentraciones de PAHs aumentaron ligeramente y este crecimiento

se mantuvo estable al final del ciclo de alimentación (últimos tres meses).

Estos resultados demuestran alta calidad del producto obtenido, con una carga de

contaminantes muy baja, probablemente como consecuencia de los piensos utilizados,

también con una carga baja de contaminantes, y sugiere la capacidad de los peces de

metabolizar y eliminar los contaminantes ingeridos en la dieta en las condiciones de cultivo

aplicadas en el experimento.

6.2.4. Referencias.

ATSDR (Agency for Toxic Substances and Disease Registry). 1995b. Toxicological Profile for

Polycyclic Aromatic Hydrocarbons (PAHs). U.S. Department of Health and Human

Services, Public Health Service, Atlanta, GA.EPA. 1987. Quality criteria for water. EPA

440/5–86-001. US Environmental Protection Agency, Washington DC.

EPA. 1993. Provisional guidance for quantitative risk assessment of polycyclic aromatic

hydrocarbons. EPA 600/R-93/089. US Environmental Protection Agency, Washington

DC.

EPA. 2000. Guidance for assessing chemical contaminant data for use in fish advisories.

EPA. 823-B-00–008. US Environmental Protection Agency, Washington DC.

IARC. 1987. IARC monographs on the evaluation of carcinogenic risks to humans,

supplement 7. Overall evaluations of carcinogenicity: an updating of IARC

monographs, vol 1-42., Vol. International Agency for Research on Cancer, Lyon.

Directiva 2000/60/CE del Parlamento Europeo y del Consejo, de 23 de octubre de 2000, por

la que se establece un marco comunitario de actuación en el ámbito de la política de

aguas.


236

Fernández-González, V.; Concha-Graña, E.; Muniategui-Lorenzo, S.; López-Mahía, P.; Prada-

Rodríguez, D. 2007. Solid-phase microextraction-gas chromatographic-tandem mass

spectrometric analysis of polycyclic aromatic hydrocarbons. Towards the European

Union water directive 2006/0129 EC. J. Chromatogr. A. 1176, 48-56.

Latimer J.S.; Zheng J. 2003 Sources, transport and fate of PAHs in the marine environment.

In: Douben PET (ed) PAH: an ecotoxicological perspective. Wiley, Chichester, UK, pp

9–33.

Menzie, C.A.; Potocki, B.B.; Santodonato, J. 1992. Ambient concentrations and exposure to

carcinogenic PAHs in the environment. Environ. Sci. Technol. 26, 1278-1284.

Nisbet, I.C.; LaGoy, P.K. 1992. Toxic equivalency factors (TEFs) for polycyclic aromatic

hydrocarbons (PAHs). Regul Toxicol Pharm 16, 290–300.

Reglamento 1881/2006/CE de la comisión de 19 de diciembre de 2006 por el que se fija el

contenido máximo de determinados contaminantes en los productos alimenticios.

Serrano, R.; Barreda, M.; Pitarch, E. 2003. Determination of Low Concentrations of

Organochlorine Pesticides and PCBs in Fish Feed and Fish Tissues from Aquaculture

Activities by Gas Chromatography with Tandem Mass Spectrometry. J. Sep. Sci. 26,

75-86.

Tsapakis, M.; Dakanali, E.; Stephanou, E.G.; Karakassis, I. 2010. PAHs and n-alkanes in

Mediterranean coastal marine sediments: Aquaculture as a significant point source. J.

Environ. Monitor. 12, 958-963.

USEPA. Toxicological profile for polyclyc aromatic hydrocarbons (PAHs).

http://www.epa.gov

Van der Oost, R.; Beyer, J.; Vermeulen, N.P.E. 2003. Fish bioaccumulation and biomarkers

in environmental risk assessment: A review. Environ. Toxicol. Pharm. 13, 57-149.

Capítulo 6. Bioacumulación de contaminantes OCs

237

6.3. Estudio de la bioacumulación de contaminantes organoclorados en doradas

procedentes de la acuicultura.


238


239


Compuestos químicos de toxicidad reconocida han sido utilizados en grandes

cantidades desde la antigüedad tanto en la agricultura como en actividades industriales

(Lacorte et al., 2006). Entre ellos, los PCBs, PCDDs y OCPs como el DDT y sus derivados DDD

y DDE son considerados sustancias de riesgo y hoy día, a pesar de estar prohibido su uso

desde hace años, es posible encontrar concentraciones de estos compuestos asociados a

muestras medioambientales y seres vivos.

Los PCBs se han utilizado comercialmente desde 1930 en numerosas aplicaciones

pero principalmente como fluidos dieléctricos e intercambiadores de calor en

transformadores y condensadores debido a su baja conductividad eléctrica, elevada

conductividad térmica y gran resistencia a la degradación por el calor. En los años 60 la

comunidad científica inició una llamada de atención respecto al uso y la peligrosidad de los

PCBs, dadas sus características de baja biodegradabilidad y el carácter bioacumulativo del

producto. Su producción está hoy prohibida en casi todos los países desarrollados,

habiéndose establecido también condiciones especiales para la utilización y destrucción de

los aparatos que contenían PCBs (Directiva 96/59/CE de 1996). La directiva 96/59/CE

establece la prohibición de su abandono y evacuación incontrolada. También regula su

eliminación controlada, regula el procedimiento de autorización de las instalaciones de

descontaminación y aplica el principio “quien contamina, paga”. Como sucede con la

mayoría de los compuestos OCs, los PCBs son muy persistentes y se acumulan en la cadena

alimentaria. A medida que aumenta el número de átomos de cloro se incrementa su

estabilidad y liposolubilidad. Los distintos congéneres se identifican con un sistema de

numeración propuesto por Ballschmiter y Zell (Ballschmiter and Zell, 1980).

Hasta hace muy pocos años no se disponía comercialmente de patrones analíticos de

congéneres individuales y la exposición a estos contaminantes se expresaba como ingesta

de PCBs totales, siendo esta aproximación insuficiente por no tener en cuenta la diferente

toxicidad de los distintos congéneres. Por otro lado, no existe un valor de referencia


240

toxicológico generalmente aceptado para la exposición a PCBs totales, comprometiendo

todo ello la evaluación de riesgos.

Los congéneres no-orto sustituidos y en menor medida algunos mono-orto

sustituidos, pueden adoptar una configuración plana que les confiere características

específicas incluyendo respuestas bioquímicas y tóxicas, algunas de las cuales, coinciden

con las causadas por la 2,3,7,8-tetraclorobibenzo-p-dioxina (Figura 6.2). Los congéneres de

PCBs que exhiben una actividad toxicológica homóloga a la de las dioxinas se denominan

Dioxin-like PCBs (DL-PCBs) o PCBs coplanares (PCB nº: 77, 105, 118, 126, 156, 169). En la

Figura 6.3. se representan tres ejemplos de DL-PCBs.

Figura 6.2. 2,3,7,8-tetraclorobibenzo-p-dioxina.

Cl

Cl Cl

Cl

PCB 77

Cl

Cl Cl

Cl

ClCl

PCB 169

Cl

Cl Cl

Cl

Cl

PCB 126

Figura 6.3. Estructura molecular de PCBs coplanares.

Cl

Cl O

O

Cl

Cl


241

Todos los PCBs cumplen una serie de características estructurales que les permiten

ser reconocidos por el receptor específico Ah-receptor (receptor para aril-hidrocarbono),

que también reconoce a otras moléculas como los PAHs (Safe et al., 1994).

Han sido establecidos TEFs para los PCBs, expresándose la toxicidad de los mismos, al

igual que para las dioxinas, en forma de equivalentes tóxicos (Van den Berg et al., 1998).

Generalmente la presencia de dioxinas en diferentes muestras analíticas hace que se suela

evaluar también la presencia de PCBs. Para evaluar la exposición a los PCBs se suelen incluir

tres PCBs no-orto sustituidos (77, 126, 169) y dos mono-orto (105, 118) para los que existen

TEFs. Los seis restantes que se suelen incluir representan algunos de los mayoritarios en

alimentos (28, 52, 101, 138, 153 y 180) (Bordajandi et al., 2006).

Por otro lado, los OCPs son un grupo de POPs que han sido utilizados en gran

medida para evitar plagas y combatir enfermedades tan graves como la malaria o la fiebre

amarilla. En muchos casos los OCPs son utilizados para combatir las plagas de insectos, los

hongos o mala hierba en superficies de cultivo. Este hecho ha dado lugar a su presencia en

el medio ambiente, llegando a la población humana. La estructura de estos compuestos

orgánicos puede ser muy diversa en la que átomos de cloro sustituyen a átomos de

hidrógeno. Aunque su uso está hoy estrictamente restringido, debido a su persistencia en el

medio ambiente, todavía se encuentran concentraciones apreciables en todos los

compartimentos del ecosistema. Como ejemplo, el DDT y sus derivados DDD y DDE son

frecuentemente detectados en el medio ambiente y asociados al tejido de algunos

animales y personas (Chu et al., 2003; Yang et al., 2006).

Tanto los PCBs como los OCPs se consideran potencialmente peligrosos por

provocar daño al sistema nervioso central y periférico, desordenes reproductivos,

trastornos al sistema inmunológico, malformaciones en el nacimiento y hasta la muerte.

Algunos de ellos poseen actividad mutagénica y carcinogénica y es muy extensa su

presencia medioambiental y en organismos vivos (Muir et al., 1999; Domingo et al., 2007).

En muchos casos son ingeridos alimentos u organismos contaminados por PCBs u OCPs.

Una de las características más importantes de los POPs es su capacidad para bioacumularse


242

en tejidos animales. Además dentro de la cadena trófica, su concentración incrementa por

cada nivel trófico, magnificándose su presencia en el hígado de los consumidores. Ha sido

comprobado en algunos casos que la acumulación de estos contaminantes en los animales

de granja está debida preferentemente, a la ingesta de piensos contaminados. Un ejemplo

claro lo representan los peces procedentes de la acuicultura. Estos muestran

contaminación por PCBs y algunos OCPs, siendo la alimentación el único factor al que se

encuentran expuestos estos animales (Smith et al., 2002; Serrano et al., 2003a). El hombre

no está exento de este peligro puesto que el consumo de alimentos procedentes de la

acuicultura marina ha crecido mucho en los últimos años y el consumo de organismos

contaminados constituye un verdadero riesgo para la salud.

En el presente estudio se determinó la concentración de OCPs y PCBs en materias

primas, piensos y filetes de dorada procedentes de un estudio experimental de cultivo de

doradas (Sparus aurata L.) en el marco del Proyecto AQUAMAX (Sustainable Aquafeeds to

Maximise the Health Benefits of Farmed Fish for Consumers. www.aquamaxip.eu). La

información generada a partir de estos estudios es de gran importancia pues permitirá

identificar cuales son las concentraciones de estos compuestos y el BMF de contaminantes

en el individuo una vez haya alcanzado su tamaño comercial.


243


Effects of fish oil replacement and re-feeding on the bioaccumulation of

organochlorine compounds in gilthead sea bream (Sparus aurata L.) of market size.

J. Nácher-Mestre, R. Serrano, L. Benedito-Palos, J. C. Navarro, F. J. López, J. Pérez-

Sanchez

Chemosphere 76 (2009) 811-817

Chemosphere 76 (2009) 811-817

244


245

Abstract

Organochlorine pesticide residues and polychlorinated biphenyls were determined in raw materials,

fish feeds and fillets from fish exposed through the productive cycle (14 months) to experimental

diets with different percentages of fish oil replacement with vegetable oils. Detectable amounts of

organochlorine compounds were found in raw materials derived from fish sources with none being

detected in vegetable ingredients. Fish feeds presented trace concentrations of contaminants at the

ng/g level, which varied according to the contribution of the different resources used in their

manufacture. Contaminants did not accumulate during the first 11 months of exposure, and low

concentrations of organochlorine compounds were found both at the start and at the end of this

feeding period. Fillets from fish fed the fish oil diet presented the highest concentrations of

organochlorine compounds, with these decreasing in proportion to fish oil replacement. Three

months of fish oil re-feeding during the finishing phase only produced significant bioaccumulation

over the course of the first month. By optimizing fish meal and fish oil replacement with vegetable

oils alternative feeds can contribute to significantly reduce the risk of organochlorine uptake by

consumers.

Keywords: Organochlorine compounds; Bioaccumulation; Sparus aurata; Fish feed; Marine

aquaculture

Chemosphere 76 (2009) 811-817

246

1. Introduction

Polychlorinated biphenyls (PCBs) and organochlorine pesticides (OCPs) are non polar, highly

lipophilic, and persistent ubiquitous environmental pollutants. Both are classified as Persistent

Organic Pollutants (POP) and are present in the contamination pattern of marine environments

worldwide. Despite the use of OCPs being strongly restricted and PCBs production being banned,

these compounds are distributed across the marine environmental biota (Hernández et al., 2000;

Hoekstra et al., 2003; Bocquene et al., 2005; Yang et al., 2007; Serrano et al., 2008b). Special concern

exists about dioxin-like PCB (DL-PCBs), which have been shown to cause toxic responses similar to

those induced by 2,3,7,8-tetrachlorodibenzo-p-dioxin, the most potent congener within

polychlorinated dibenzo-p-dioxins (Van Den Berg et al., 1998). The characteristics of these

compounds lead to high biomagnification in the food chain, and involve a wide range of trophic

levels (Borga et al., 2001; Kidd et al., 2001; Hoekstra et al., 2003; Konwick et al., 2006; Serrano et al.,

2008a,b; Sagratini et al., 2008). Some studies have shown that food is the major contributor for PCBs

accumulation in farmed fish (Serrano et al., 2003a; Antunes and Gil, 2004). Thus, fish and in general

seafoods have been considered the most important source of organochlorine compounds (OCs) in

the human diet (Johansen et al., 1996; Bjerregaard et al., 2001; Tsukino et al., 2006), these

compounds being frequently detected in human tissue lipids and fluids (Hernández et al., 2002a,b,c;

Pitarch et al., 2003; De Felip et al., 2004; Muñoz-de-Toro et al., 2006; Tsukino et al., 2006; Lopez-

Espinosa et al., 2008; Mueller et al., 2008).

Persistent organic pollutants are concentrating in lipid-rich feed grade fish used for

production of fish oils, the major resource of these contaminants in aquaculture diets, and have

been detected in fish feed used in aquaculture and in farmed fish by several authors (Santerre et al.,

2000; Easton et al., 2002; Hites et al., 2004; Navas et al., 2005; Bordajandi et al., 2006; Maule et al.,

2007; Ábalos et al., 2008; Serrano et al., 2008a,b), so dietary fish oil replacement with alternative oils

can significantly reduce the load-charge of lipophilic contaminants in aquafeeds and thereby farmed

fish (Bell et al., 2005; Berntssen et al., 2005; Bethune et al., 2006). Unfortunately, when fish oil is

replaced with vegetable oils, the dietary supply of n-3 polyunsaturated fatty acids (PUFA) is also

reduced (Bell et al., 2005; Benedito-Palos et al., 2007, 2008; Drew et al., 2007). This is a major

constraining factor for marine fish due to the inability of marine fish to convert C18 PUFA to long-

chain PUFA, specially eicosapentaenoic acid (20:5 n-3) and docosahexaenoic acid (22:6 n-3) that

become essential nutrients in marine fish aquafeeds (Sargent et al., 1999, 2002). However, among

others, recent studies in gilthead sea bream (Sparus aurata L.), indicate that fish oil can be replaced


247

up to 66% with vegetable oils in plant protein-based diets without signs of growth retardation and

histopathological tissue damage in a 8-month trial (Benedito-Palos et al., 2007, 2008). Thus, a large

amount of either fish meal or fish oil can be, and is practically replaced in the new sustainable diets

for marine fish.

The feasibility of the fish oil replacement has been demonstrated through an entire

production cycle, including a three month finishing fish oil diet phase (wash-out period), without

negative effects on the growth performance of gilthead sea bream, a high valuable fish for the

Mediterranean aquaculture. Also, a kinetic analysis of the fatty acids (FA) demonstrated that

changes in the fillet FA profile arise because the existing stores become diluted as fish grow and

deposit increasing amounts of dietary derived FAs (Benedito-Palos et al., 2009). The goal of the

present study was to analyze, in the same fish, OCPs and PCBs bioaccumulation. Organochlorine

compounds were determined in fish fillets, feeds and major raw materials of the diets, providing

information on the end-product quality and safety from the consumer point of view.

2. Materials and methods

2.1. Experimental diets

Three isoproteic, isolipidic and isoenergetic plant protein-based diets were formulated with a

low inclusion level (20%) of fish meal and fish soluble protein concentrates. Fish oil from the

southern hemisphere was the only lipid source in the control diet (FO), which was also used as a

finishing diet. The two remaining diets contained a blend of vegetable oils (2.5 rapeseed oil:8.8

linseed oil:3 palm oil), replacing 33% (33VO) and 66% (66VO) of the fish oil. All diets were

manufactured using a twin-screw extruder (Clextral, BC 45) at the “Institut Scientifique de Recherche

Agronomique” (INRA) experimental research station of Donzacq (Landes, France), dried under hot

air, sealed and kept in air-tight bags until use. Ingredients and proximate composition are shown in

Table 1.

Samples for pollutant analyses were collected and stored at −20 °C until analysis.

Determination of organochlorine residue and polychlorinated biphenyls in each diet was carried out

by triplicate.

Chemosphere 76 (2009) 811-817

248

Table 1.

Ingredients and chemical composition of experimental diets. For more details in diet composition see Benedito-Palos et al., 2009.

Ingredient (%) FO 33VO 66VO

Fish meal (CP 70%) a

15 15 15

CPSP 90 b

5 5 5




Fish oil c

15.15 10.15 5.15



Palm oil 0 1.25 2.5

Soya lecithin 1 1 1

Binder 1 1 1

Mineral premix d

1 1 1

Vitamin premix e

1 1 1

CaHPO4.2H2O (18%P) 2 2 2

L-Lys 0.55 0.55 0.55

Proximate composition

Dry matter (DM, %) 93.13 92.9 92.77

Protein (% DM) 53.2 52.81 52.62

Fat (% DM) 21.09 21 20.99

Ash (% DM) 6.52 6.69 6.57

aFish meal (Scandinavian LT)

bFish soluble protein concentrate (Sopropêche, France)

cFish oil (Sopropêche, France)

dSupplied the following (mg · kg /diet, except as noted): calcium carbonate (40% Ca) 2.15 g,

magnesium hydroxide (60% Mg) 1.24 g, potassium chloride 0.9 g, ferric citrate 0.2 g, potassium iodine 4 mg, sodium chloride 0.4 g, calcium hydrogen phosphate 50 g, copper sulphate 0.3, zinc sulphate 40, cobalt sulphate 2, manganese sulphate 30, sodium selenite 0.3. eSupplied the following (mg · kg /diet): retinyl acetate 2.58, DL-cholecalciferol 0.037, DL- tocopheryl

acetate 30, menadione sodium bisulphite 2.5, thiamin 7.5, riboflavin 15, pyridoxine 7.5, nicotinic acid 87.5, folic acid 2.5, calcium pantothenate 2.5, vitamin B12 0.025, ascorbic acid 250, inositol 500, biotin 1.25 and choline chloride 500.


249

2.2. Bioaccumulation experiment

Gilthead sea bream of Atlantic origin (Ferme Marine de Douhet, Ile d’Oléron, France) were

cultured at the Instituto de Acuicultura de Torre de la Sal (IATS) for 20 days before the start of the

study. Fish of around 18 g initial mean body weight were allocated into nine fiberglass tanks (3000 L)

in groups of 150 fish per tank. Water flow was 20 L/min and oxygen content of outlet water

remained higher than 85% saturation. The growth study was undertaken over 14 months (July 11th,

2006–September 2nd, 2007), and day-length and water temperature (10–26 °C) varied over the

course of the trial to match natural sea changes offshore from IATS (40° 5′N; 0° 10′E).

During the first 11 months of the trial, the three diets were randomly allocated to triplicate

groups of fish, and feed was offered by hand to apparent visual satiety. During the finishing diet

phase (12 weeks, June 6th, 2007–September 2nd, 2007), two tanks of 33VO and two of 66VO groups

were fed with FO diet. The names of those groups became 33VO/FO and 66VO/FO, respectively. Fish

fed FO diet, and one tank of fish fed 33VO and 66VO diets were maintained on the initial diets until

the end of the experiment. This meant that the bioaccumulation study consisted of five treatments:

FO, 33VO, 66VO, 33VO/FO and 66VO/FO (Fig. 1).

At the beginning and at regular intervals through the finishing diet phase (0, 330, 360, 390

and 420 days) randomly selected fish (eight fish from all tanks of the same treatment) were

sacrificed by a blow on the head prior to tissue sampling. The left-side fillet (with skin and bone

removed) was excised and stored at −20 °C until analysis. As reported by Benedito-Palos et al.

(2009), body weight and fillet yield were not affected by the dietary treatment over the course of all

feeding trial.

Sea water for fish culture was analyzed using the multi-residue method for pesticides

described by Hernandez et al. (1993) and OCs were not detected (limit of detection between 0.01

and 0.1 μg/L). Therefore, fish were cultured in sea water free of pesticides without any other known

exposure to organochlorines except feed.

2.3. Analytical methodology

Organochlorine compounds, including selected non polar pesticides and derivatives (DDTs-

p,p′-DDT, p,p′-DDE, p,p′-DDD–, HCB, lindane, mirex, methoxychlor) and polychlorinated biphenyls

IUPAC nos 28, 31, 52, 77, 101, 105, 118, 126, 128, 138, 153, 156, 169, 170, 180, as indicators of the

Chemosphere 76 (2009) 811-817

250

presence of PCBs in the samples, were analyzed in raw materials, fish feeds and fish fillets following

the method described by Serrano et al. (2003b).

July 2006 Sept 2007June 2007

Fig. 1. Schematic representation of the long-term feeding trial over 14 months (for abbreviations see the text).

Eight fillets from each treatment (four from each replicate) were selected randomly to obtain

three composite samples and were analyzed independently in triplicate. Fish diets and raw materials

were homogenized and analyzed in triplicate. In brief, extraction of fish feed, raw materials and

muscle was carried out by refluxing ca. 8 g homogenized fresh sample in n-hexane for 4 h. Clean up

of fish tissue was performed by means of normal phase liquid chromatography (NPLC) injecting 1 mL

hexanic extract (4 g sample per mL) into a silicagel HPLC column. The liquid chromatography (LC)

system used was comprised of a LC Pump Master 305 piston pump, Gilson (Middleton, USA), with a

column 150 × 3.9 mm i.d. packed with 4 μm silica Novapack (Waters, Milford, MA, USA); mobile

phases were n-hexane and ethyl acetate for column regeneration after a clean up cycle; flow rate

was 1 mL/min. For the fish feed and raw materials an additional clean-up step with concentrated


251

sulphuric acid was necessary prior to normal phase HPLC clean up as a consequence of the high lipid

content (from 20% to 100% of fresh weight) (Serrano et al., 2003b).

Analysis of the fat-free LC fractions collected was performed by gas chromatography with

tandem mass spectrometry (GC–MS/MS) using a gas chromatograph Varian CP-3800 coupled with an

ion trap mass spectrometry detector (Saturn 4000, Varian) operating in the electron impact

ionisation mode (EI) to identify and quantify PCBs and OCPs present in the diets and fish fillets.

Separation of the analytes was carried out on a 30 m × 0.25 mm DB-5MS (0.25 μm film thickness)

Varian capillary column, using helium at 1 mL/min as the carrier gas. The temperature program was

as follows: 90 °C for 1 min, increased to 180 °C at a rate of 30 °C/min, increased to 260 °C at a rate of

4 °C/min, and finally increased to 300 °C at a rate of 20 °C/min, with a final isothermal stage of 4 min

(total chromatographic analysis time was 30 min). Injection of 1 μL of samples (injection port

temperature 250 °C) in the splitless mode was carried out using a Varian 8400 autosampler equipped

with a 10 μL syringe. The gas chromatograph was directly interfaced with the Varian 4000 mass-

spectrometer (ion trap) in the internal ionization mode with electron impact ionization energy of 70

eV in the positive ion mode. Transfer line temperature was established at 260 °C and ion source and

trap temperatures were adjusted to 200 °C (more details in Serrano et al. (2003b) Quantification of

samples was carried out by external calibration curve using pp′-DDE-D8 as internal standard. Fifty

nanograms of this isotopically labeled standard (100 μL of a 0.5 ng/μL hexanic standard solution)

were added before extraction, allowing its use as internal standard and surrogate in order to detect

possible analytical errors in every sample analysis. The limit of quantification has been calculated as

three times the detection limit (i.e. nine times the background noise in the chromatograms),

accepting coefficients of variation less than 30%. Statistical validation of the method included the

study of linearity, accuracy, precision, selectivity and limits of quantification and detection.

Moreover, the absence of matrix effect in the instrumental determination (GC–MS/MS) was proved

by means of the Standard Additions Method (more details Serrano et al., 2003b).

The whole analytical process was carried out in the Good Laboratory Practices certified

laboratories of the Research Institute for Pesticides and Water, University Jaume I, Spain.

2.4. Determination of fat

The total fat content in the sample extracts was determined gravimetrically, with evaporation

at 95 °C until constant weight.

Chemosphere 76 (2009) 811-817

252

2.5. Data analysis

PCBs and OCPs concentrations in raw materials and fish feed are expressed as ng/g fresh

weight and in fillets also as lipid based concentrations. Biomagnification Factors (BMF) were

calculated as the ratio between lipid based concentrations of organochlorine compounds in the fish

fillets and in the fish feed at each sampling point. As two different fish feeds were used in 33VO/FO

and 66VO/FO groups, calculations of BMF were made taking into account the time of exposure to

each fish feed. Thus, the arithmetic mean of organochlorine concentrations in the diet was

calculated on both the basis of months of exposure to each fish feed and their concentration in each

fish feed. The calculated data expressed as ng/fillet allowed to identify the amount of pollutant

present in a whole fish fillet. The t-Student test (P < 0.05) was applied to compare data on diets and

ingredients. Data for concentrations of PCBs and OCPs in fish fillets, BMF values and amounts of

pollutants per fillet were compared by means of ANOVA I and “a posteriori” Scheffe´s test (P < 0.05).

All data were transformed to Log 10 before statistical analysis to achieve normality.

Homoscedasticity of variances was tested by means of Bartlett’s test (P < 0.05). All the statistical

tests were conducted using STATGRAPHICS plus for Windows version 4.1 (Statistical Graphics

Corporation).

3. Results and discussion

3.1. PCBs and OCPs content in experimental diets

The analytical method showed excellent sensitivity and selectivity as a consequence of the

use of gas chromatography coupled tandem mass spectrometry (GC–MS/MS). It was validated by

recovery experiments down to 5 ng/g, with lower spiked levels not being checked due to the

presence of DDTs and PCBs in the blank samples (Serrano et al., 2003b). However, the powerful

analytical characteristics of GC–MS/MS together with the efficiency of the HPLC clean up has allowed

here analyte quantification at concentrations as low as 0.1 ng/g with acceptable precisions (below

30%). Samples analysis were carried out in the Research Institute for Pesticides and Water facilities,

whose laboratories are Good Laboratory Practices certified, which support the quality of the

analytical data released in the present research.


253

Table 2. Concentration (ng/g fresh weight) of organochlorine compounds in marine resources used in fish feed manufacture and experimental diets. Pollutants in vegetable resources were no detected.

Compound Fish oil Fish meal Fish soluble protein concentrate Diet FO Diet 33VO Diet 66VO

Mean CV Mean CV Mean CV Mean CV Mean CV Mean CV

PCB 28+31 1.6 4 <0.1 – <0.1 – 0.2 9 0.1 9 <0.1

PCB 52 1.2 14 <0.1 – <0.1 – 0.2 6 0.2 8 0.1 15

PCB 101 3.6 1 0.7 29 0.5 6 0.5 6 0.5 9 0.3 13

PCB 118 4.7 9 <0.4 – 1 1 0.6 5 0.6 5 0.4 3

PCB 153 6.5 19 2.1 4 3.5 14 1.6 3 1.5 6 0.9 7

PCB 105 3.8 5 <0.2 – 0.3 22 <0.1 3 <0.1 – <0.1 –

PCB 138 4.9 8 1.4 20 1.9 3 0.8 14 0.7 7 0.5 4

PCB 128 2 7 <0.2 – 0.3 27 <0.1 – <0.1 – <0.1 –

PCB 156 1.9 4 <0.4 – <0.4 – <0.1 – <0.1 – <0.1 –

PCB 180 4.8 14 0.4 14 0.9 12 0.4 8 0.4 37 0.3 11

PCB 170 0.6 31 0.3 11 0.3 18 <0.2 – <0.1 – <0.1 –

SUM PCBs 34.8 4 4.7 18 8.7 11 4.4 3 4.1 5 2.8 3

HCB <0.1 – <0.1 – <0.1 – 0.2 18 0.2 30 0.5 4

p,p-DDE 16.8 3 5.5 7 3.6 7 2.5 12 2 10 1.7 5

p,p-DDD 8.8 29 1.4 7 0.6 15 2.3 4 2 1 1.3 3

p,p-DDT 15.9 6 1.5 12 1 11 2.5 10 2.1 2 1.4 4

SUM OCPs 41.6 7 8.4 6 5.2 2 7.6 5 6.2 3 4.8 2

Total load 76.4 4 13.1 5 13.9 4 11.9 3 10.2 2 7.6 2

CV: coefficient of variation. Compounds were quantified with CV < 30.

Chemosphere 76 (2009) 811-817

254

Concentrations of PCBs and OCPs present in the raw materials used for the manufacture of

experimental diets are shown in Table 2. As can be observed raw materials derived from fish contain

detectable amounts of organochlorine compounds, up to 76.4 ng/g of total organochlorine charge-

load (34.8 ng/g total PCBs and 41.6 ng/g total OCPs). The predominant pesticide was pp′-DDT and its

derivatives. For the PCBs investigated, the analytical methodology applied allowed the detection and

quantification of 12 congeners from 15 considered as markers of total load. Congener 153 was the

most abundant, as is usual in marine samples (Serrano et al., 2008a,b). This pattern is in agreement

with the pollution pattern of different marine products (Zabik et al., 1996; Santerre et al., 2000;

Johnson et al., 2007; Serrano et al., 2008a, 2008b).

The presence of organochlorine contaminants in fish oil has been reported previously by

other authors. Jacobs et al. (1997, 1998) reported concentrations up to 990 μg/L total PCBs and 119

μg/L total DDTs. Hilbert et al. (1998) reported concentrations up to 300 ng/g total PCBs and 60 ng/g

total DDTs during the refining process. In all cases, levels reported are higher than those found in the

raw materials used in this work. This fact could be related with the northern origin of the raw

materials used in the cited papers.

Organochlorine compounds were not detected in any vegetable ingredient used to formulate

the experimental diets, which allows the total load of contaminants in diets to be reduced with fish

oil replacement. The non detectable levels of organochlorine contaminants in vegetable derivative

resources and the low levels present in the marine derivatives used allow production of a new

generation of fish feeds almost free of contaminants.

Table 2 also shows the concentrations of OCPs and PCBs present in the experimental diets

used during the diet exposure experiment. The sum of OCPs only reached 7.6 ng/g fresh weight in

the case of the FO diet, the diet with highest fish oil content. The sum of PCBs analyzed reached 4.4

ng/g fresh weight in the reference fish feed (FO).

The charge-load of contaminants present in the diets is in accordance with the contribution of

the different resources used in their manufacture (t-Student, P < 0.05), which leads to very low levels

of organochlorines. Thus fish oil replacement by vegetable oils without detectable levels of

contaminants, decreases the charge of contaminants in correlation with the percentage of fish oil in

the different diets (r = 0.9978, P < 0.01). Similar results have been previously obtained by other

authors (Bell et al., 2005; Berntssen et al., 2005).


255

The concentrations of organochlorines in different fish feeds have been reported by several

authors (Easton et al., 2002; Hites et al., 2004; Carubelli et al., 2007; Maule et al., 2007) with loads of

contaminants in the different feeds higher or similar than levels found in the experimental diets used

in this work. This highlights the quality of the new diets employed in this study in comparison to

others. It is interesting to note that feeds for salmon farming present higher levels of contaminants

than those used for Mediterranean aquaculture (Hites et al., 2004). This fact could be related to the

use of fish oil from northern Atlantic and the higher percentage of fish oil used in salmonid diets

(around 25–33% versus 15–20% in Mediterranean aquaculture feeds), or other factors such as the

quality of raw materials used in manufacture.

3.2. Fish diet exposure

Table 3 and Table 4 show the concentrations of PCBs and OCPs in fish fillets during the long

term diet exposure experiments. Fish fed the FO diet presented the highest concentrations of

organochlorine compounds at the end of the experiment (PCBs and OCPs) (Scheffe’s test, P < 0.05)

as would be expected due to the higher content of pollutants quantified in this diet. Following the

expected trend, the fish fed 33VO presented lower organochlorine concentrations than those fed FO

diet but were higher than 66VO group (P < 0.05). For the experiments with finishing diets, 33VO/FO

and 66VO/FO, the supply of FO diet during the last three months of the experiment, provokes a

tendency towards the increase of the levels of organochlorine compounds in the fish fillets with

respect to fish kept without finishing diets (33VO and 66VO), although not reaching the levels of

PCBs and OCPs found in the control FO group.

Table 3 and Table 4 also show the trend of PCBs and DDTs levels during the bioaccumulation

experiment. During the first 11 months, contaminant concentrations increased very little compared

with the last three months of the experiment. In fact contaminants did not accumulate during the

first 11 months of exposure to the experimental diets, presenting similar concentrations (referred to

both fresh and lipid weight) of organochlorine compounds at the start and finish of this feeding

period (fish mean weights were 18 g at the start and 290 g after 11 months of exposure) (sampling

times 0 and 1, July 06 and June 07, respectively) (P > 0.05). The fact that levels of PCBs, dioxin-like

Chemosphere 76 (2009) 811-817

256

Table 3. Concentrations of total PCBs and total DL-PCBsa in fillets (mean ± coefficient of variation, n = 3) from fish exposure to the different experimental diets in long-term sea bream feeding trial (see Fig. 1).

Compound

T0

FO

33VO

33VOFO

66VO

66VOFO

July/06 June/07 Sept/07 June/07 Sept/07 Sept/07 June/07 Sept/07 Sept/07

ng/g fresh weight

PCB 28+31 <0.1 <0.1 0.2 (19) <0.1 0.1 (26) 0.2 (19) <0.1 0.2 (4) 0.2 (18)

PCB 52 <0.1 0.1 (27) 0.3 (15) <0.1 0.2 (35) 0.2 (35) <0.1 0.2 (9) 0.2 (21)

PCB 101 <0.1 0.2 (30) 0.6 (31) 0.1 (22) 0.4 (27) 0.4 (31) <0.1 0.3 (16) 0.5 (15)

PCB 77 <0.2 <0.2 <0.2 <0.2 <0.2 <0.2 <0.2 <0.2 <0.2

PCB 118 0.5 (9) 0.5 (13) 1 (16) 0.2 (23) 0.7 (31) 0.8 (16) 0.3 (25) 0.6 (12) 0.8 (2)

PCB 153 2 (26) 0.7 (19) 1.8 (26) 0.4 (25) 1.4 (29) 1.6 (26) 0.5 (28) 1.2 (12) 1.8 (14)

PCB 105 <0.2 0.5 (30) 0.6 (25) <0.2 0.3 (18) 0.4 (25) 0.2 (19) 0.3 (21) 0.4 (13)

PCB 138 1 (11) 0.6 (8) 1.1 (22) 0.2 (22) 0.8 (29) 0.9 (22) 0.4 (23) 0.7 (16) 1.0 (5)

PCB 126 <0.2 <0.2 <0.2 <0.2 <0.2 <0.2 <0.2 <0.2 <0.2

PCB 128 <0.2 0.2 (4) 0.3 (19) 0.1 (20) 0.2 (14) 0.2 (19) 0.1 (24) 0.2 (10) 0.2 (24)

PCB 156 <0.4 <0.4 <0.4 <0.4 <0.4 <0.4 <0.4 <0.4 <0.4

PCB 180 0.5 (26) <0.1 <0.1 <0.1 <0.1 <0.1 <0.1 <0.1 <0.1

PCB 169 <0.1 0.5 (33) 0.6 (24) 0.2 (30) 0.5 (19) 0.5 (24) <0.1 0.4 (16) 0.6 (23)

PCB 170 <0.2 0.3 (3) 0.2 (23) 0.1 (6) 0.2 (23) 0.1 (23) 0.1 (4) 0.2 (14) 0.1 (20)

SUM PCBs 4.0 (3) 3.7 (11) 6.5 (5) 1.7 (10) 4.8 (11) 5.4 (11) 1.6 (14) 4.1 (5) 5.7 (6)

SUM DL-PCBsa 0.4 (1) 1.2 (0.2) 1.3 (0.1) 0.5 (0.1) 1.0 (0.1) 1.0 (0.3) 0.5 (0.1) 0.8 (0.1) 1.1 (0.2)

ng/g lipid weight

SUM PCBs 83.5 (4) 47.5 (2) 70.6 (6) 22.6 (2) 48.3 (5) 51.3 (5) 27.9 (4) 40.6 (2) 51.9 (3)

SUM DL-PCBsa 9.8 (1) 15.4 (2) 14.6 (2) 7.1 (1) 10.2 (1) 9.7 (2) 6.9 (1) 8.5 (1) 10.1 (2)


257

a See list of PCBs analyzed in Section 2.

Table 4. Concentrations of total OCPs in fillets (mean ± coefficient of variation, n = 3) from fish exposure to the different experimental diets in long-term sea bream feeding trial (see Fig. 1).

Compound

T0

FO

33VO

33VOFO

66VO

66VOFO

July/06 June/07 Sept/07 June/07 Sept/07 Sept/07 June/07 Sept/07 Sept/07

ng/g fresh weight

HCB <0.1 <0.1 0.2 (22) <0.1 <0.1 0.2 (34) <0.1 0.4 (2) <0.1

p,p-DDE 2.2 (5) 2(15) 3.6 (19) 1.0 (8) 2.5 (20) 2.6 (30) 0.7 (18) 2.1 (11) 2.0 (5)

p,p-DDD 0.7 (11) 1.4 (19) 2.3 (15) 0.4 (29) 1.7 (29) 2.1 (28) 0.6 (10) 1.3 (8) 1.8 (4)

p,p-DDT <0.4 1.2 (24) 2.2 (13) 0.3 (29) 1.4 (19) 1.6 (28) 0.6 (33) 1.0 (19) 1.8 (4)

SUM OCPs 2.9 (5) 4.6 (11) 8.3 (4) 1.7 (11) 5.8 (7) 6.6 (7) 1.8 (12) 4.8 (5) 5.9 (4)

ng/g lipid weight

SUM OCPs 67.1 (1) 58.9 (6) 90.4 (3) 22.7 (2) 58.5 (4) 62.6 (4) 23.6 (3) 47.8 (2) 54.1 (2)

Chemosphere 76 (2009) 811-817

258

PCBs and organochlorine pesticides in fish fillets did not increase in fresh weight and lipid based

concentrations indicates the no bioaccumulation of these contaminants, suggesting the ability of fish

to depurate the pollutants ingested in the period previous to the start of the experiment as well as in

the first 11 months. Possibly dilution, excretion or detoxifying systems operate efficiently when

toxicant load is low in diets with adequate nutritional composition.

From June 2007 to September 2007 fish ingested almost the same weight of feed as in the

previous experimental period (from July 2006 to June 2007) (Fig. 1). An increase in food intake

during the finishing diet phase can cause major increases in the concentration of pollutants. Such

increases can be seen after June 2007. As consequence of the different feeding behavior during the

grow out and finishing periods, a kinetic model study was not considered. Treatments 33VO/FO and

66VO/FO showed a higher change compared to fish always fed FO, 33VO and 66VO diets. The

ingestion of FO diet from June 2007 onwards as well as the higher feed intake resulted in a major

increase in pollutants charge in the fish fillets. In a short time fish accumulated more pollutants from

the diet than could be eliminated by any detoxification route.

Fig. 2 shows the Biomagnification Factors (BMF) at the four sampling points during the last

three months of total experimental period. As can be observed, BMF values were close to one at

sampling point 1 (July 07) and, fish fed on the diet with no fish oil replacement presented a steady,

continuous increase over time. This same trend was observed but was softer in fish fed always 33VO

and 66VO diets. Fish submitted to fish oil re-feeding showed stable BMF values (Scheffe’s test, P >

0.05) the three last sampling points. This observation supports the hypothesis discussed above on

the high efficacy of detoxification routes, probably due to low oxidative stress because of the very

low levels of contaminants and adequateness of the fish feeds composition supplied.

In September 2007 the total feeding trial finished as fish had reached commercial size. The

concentration of the sum of PCBs and OCPs in fish fillets at this time revealed the low charge of

contaminants and allowed the contaminant intake per fish by consumers to be checked (Fig. 3). As

shown, the charge of pollutants per fillet decreased with reduced percentages of fish oil in their diet.

As well, experimental diets that included FO as finishing diets for the three last months did not

increase the load of contaminants up to the load present in fish fed FO for the full 14 months

(Scheffe’s test, P > 0.05). Besides, polybrominated biphenyl diethyl ethers (PBDEs) have been

analyzed in fish feed and fish fillets studied in this work, showing very low levels of individual PBDEs


259

around 0.2–0.3 ng/g fresh weight in fish feeds and below 0.1 ng/g fresh weight in fish fillets (data not

shown).

Several authors have reported the presence of persistent organic contaminants in cultured

fish. Santerre et al. (2000) noticed that farmed channel catfish, rainbow trout and red swamp

crayfish from Southern USA sources presented absence or low levels of organochlorine

contaminants, which were lower than the concentration levels reported in wild fish. On the other

hand, Easton et al. (2002) assessed the risk of the organochlorine contaminants in wild and farmed

salmon, indicating higher levels of PCBs and OCPs in cultured fish as a consequence of the presence

of these contaminants in fish feed.

Fig. 2. Biomagnification Factors for total OCs (ng/g lipid based) of fish fed different experimental diets with varying levels of fish oil substitution, at four sampling points (Fig. 1 for details). Letters: comparison among BMF at different sampling points for each experimental diet, Scheffe’s test, P < 0.05. bars: standard deviation.

0

0,5

1

1,5

2

2,5

3

FO 33VO 33VO/FO

experimental diets

BM

F (

tota

l OC

slip

id b

ased

)

time 1 time 2 time 3 time 4

0

0,5

1

1,5

2

2,5

3

FO 66VO 66VO/FO

experimental diets

BM

F (

tota

l OC

slip

id b

ased

)


0

0,5

1

1,5

2

2,5

3

FO 33VO 33VO/FO

experimental diets

BM

F (

tota

l OC

slip

id b

ased

)


0

0,5

1

1,5

2

2,5

3

FO 66VO 66VO/FO

experimental diets

BM

F (

tota

l OC

slip

id b

ased

)


a

b

b

c

a

b b

b

a

b b

a

b

b

b

a

b

c

b

a

b

b

b

a

bc

b

a

b

a

b

b

a

b

c

b

a

b

c

b

a

b

Chemosphere 76 (2009) 811-817

260

This fact has been confirmed in a recent study carried out by Hites et al. (2004), also

suggesting the fish feed consumed is the cause of the high levels of contaminants in farmed salmon,

especially in Northern Europe. In these samples, the levels of organic compounds were higher than

those found in farmed fish from North and South America.

In regard to temperate marine aquaculture, (Serrano et al., 2008a) and (Serrano et al., 2008b)

reported higher levels of contaminants in wild fish than those found in cultured fish from the

Fig. 3. Concentration (ng/fillet fresh weight) of sum PCBs and OCPs in fillets of fish fed different experimental diets with varying levels of fish oil substitution at the end of the long-term feeding trial (September 2007). Fillet weight was not altered by dietary intervention and ranged between 113 and 126 g (Benedito-Palos et al., 2009). Letters: comparison among diets for every category (Capitals: sum PCBs, Lower case: sum OCs). Scheffe’s test, P < 0.05. bars: standard deviation.

Western Mediterranean region. This could be explained again by of the low contaminants load of

fish feeds supplied to the farmed fish (see Table 2). From this point of view and in a food-safety

framework, the results of the present work show that there is further margin for improvement in the

elaboration of “cleaner” feeds and cultured fish.

1200

1000

800

600

400

200

0

FO 33VO 66VO 33VO/FO 66VO/FO

SUM PCBs SUM

A

CC

BB

a

cc

b

b

1200

1000

800

600

400

200

0


SUM PCBs SUM OCPs

A

CC

BB

a

cc

b

b

Concentration(ng·fillet-1

1200

1000

800

600

400

200

0


SUM PCBs SUM

A

CC

BB

a

cc

b

b

1200

1000

800

600

400

200

0


SUM PCBs SUM OCPs

A

CC

BB

a

cc

b

b

Concentration(ng/filletfreshweight) 1200

1000

800

600

400

200

0


SUM PCBs SUM

A

CC

BB

a

cc

b

b

1200

1000

800

600

400

200

0


SUM PCBs SUM OCPs

A

CC

BB

a

cc

b

b


1200

1000

800

600

400

200

0


SUM PCBs SUM

A

CC

BB

a

cc

b

b

1200

1000

800

600

400

200

0


SUM PCBs SUM OCPs

A

CC

BB

a

cc

b

b

1200

1000

800

600

400

200

0


SUM PCBs SUM

A

CC

BB

a

cc

b

b

1200

1000

800

600

400

200

0


SUM PCBs SUM OCPs

A

CC

BB

a

cc

b

b


1200

1000

800

600

400

200

0


SUM PCBs SUM

A

CC

BB

a

cc

b

b

1200

1000

800

600

400

200

0


SUM PCBs SUM OCPs

A

CC

BB

a

cc

b

b

Concentration(ng/filletfreshweight)


261

Levels of pollutants in initial fish (T0) (Table 3 and Table 4), can be attributed to the nutritional

background of the previous diets that influenced the profile of pollutants. These were commercial

diets with major levels of pollutants compared to the new generation of diets used in this work.

Besides, it has to be taken into account that the maternal transfer of organochlorine compounds in

fish has been reported as a possible source of accumulation (Miller, 1993; Russell et al., 1999;

Monosson et al., 2003; Oka et al., 2006; Serrano et al., 2008c).

In this study fillets from fish fed FO diet presented trace level concentrations of

organochlorine compounds with this decreasing in proportion to fish oil replacement. This fact has

made it possible to produce a new generation of aquaculture fish feeds with optimal replacement of

fish meal and fish oil with sustainable, alternative feed resources such as vegetable oils and meals. In

this way, feeds can be produced that contribute to ensure production of healthy, marketable fish,

ensure also that the risk of organochlorine transfer to consumers is reduced, and improve

sustainability of marine aquaculture by reducing the use of wild fish sources for feed production.

Acknowledgments

This paper has been carried out with the financial support of the European Union Project:

Sustainable Aquafeeds to Maximise the Health Benefits of Farmed Fish for Consumers (AQUAMAX).

Contract Number: 016249-2.

References

Antunes, P., Gil, O., 2004. PCB and DDT Contamination in Cultivated and Wild Sea Bass from Ria De

Aveiro, Portugal. Chemosphere. 54, 1503-1507.

Bell, J.G., McGhee, F., Dick, J.R., 2005. Dioxin and Dioxin-Like Polychlorinated Biphenyls (PCBs) in

Scottish Farmed Salmon (Salmo Salar): Effects of Replacement of Dietary Marine Fish Oil with

Vegetable Oils. Aquaculture. 243, 305-314.

Benedito-Palos, L., Navarro, J.C., Bermejo-Nogales, A., 2008a. The Time Course of Fish Oil Wash-Out

Follows a Simple Dilution Model in Gilthead Sea Bream (Sparus Aurata L.) Fed Graded Levels

of Vegetable Oils. Aquaculture.

Benedito-Palos, L., Navarro, J.C., Sitjà-Bobadilla, A., 2008b. High Levels of Vegetable Oils in Plant

Protein-Rich Diets Fed to Gilthead Sea Bream (Sparus Aurata L.): Growth Performance, Muscle

Fatty Acid Profiles and Histological Alterations of Target Tissues. British J.Nutr. 100, 992-1003.

Chemosphere 76 (2009) 811-817

262

Benedito-Palos, L., Saera-Vila, A., Calduch-Giner, J., 2007. Combined Replacement of Fish Meal and

Oil in Practical Diets for Fast Growing Juveniles of Gilthead Sea Bream (Sparus Aurata L.):

Networking of Systemic and Local Components of GH/IGF Axis. Aquaculture. 267, 199-212.

Berntssen, M.H.G., Lundebye, A. Torstensen, B.E., 2005. Reducing the Levels of Dioxins and Dioxin-

Like PCBs in Farmed Atlantic Salmon by Substitution of Fish Oil with Vegetable Oil in the Feed.

Aquacult. Nutr. 11, 219-231.

Bethune, C., Seierstad, S.L., Seljeflot, I., 2006. Dietary Intake of Differently Fed Salmon: A Preliminary

Study on Contaminants. Eur. J. Clin. Investig. 36, 193-201.

Bjerregaard, P., Dewailly, E., Ayotte, P., 2001. Exposure of Inuit in Greenland to Organochlorines

through the Marine Diet. J. Toxicol. Env. Health. 62, 69-81.

Bocquené, G., Franco, A., 2005. Pesticide Contamination of the Coastline of Martinique. Mar. Pollut.

Bull. 51, 612-619.

Bordajandi, L.R., Maran, I., Abad, E., 2006. Organochlorine Compounds (PCBs, PCDDs and PCDFs) in

Seafish and Seafood from the Spanish Atlantic Southwest Coast. Chemosphere. 64, 1450-1457.

Borga, K., Gabrielsen, G.W., Skaare, J.U., 2001. Biomagnification of Organochlorines Along a Barents

Sea Food Chain. Environ. Pollut. 113, 187-198.

Carubelli, G., Fanelli, R., Mariani, G., 2007. PCB Contamination in Farmed and Wild Sea Bass

(Dicentrarchus Labrax L.) from a Coastal Wetland Area in Central Italy. Chemosphere. 68,

1630-1635.

De Felip, E., Di Domenico, A., Miniero, R., 2004. Polychlorobiphenyls and Other Organochlorine

Compounds in Human Follicular Fluid. Chemosphere. 54, 1445-1449.

Drew, M.D., Ogunkoya, A.E., Janz, D.M., 2007. Dietary Influence of Replacing Fish Meal and Oil with

Canola Protein Concentrate and Vegetable Oils on Growth Performance, Fatty Acid

Composition and Organochlorine Residues in Rainbow Trout (Oncorhynchus Mykiss).

Aquaculture. 267, 260-268.

Easton, M.D.L., Luszniak, D., Von der Geest, E., 2002. Preliminary Examination of Contaminant

Loadings in Farmed Salmon, Wild Salmon and Commercial Salmon Feed. Chemosphere. 46,

1053-1074.

Hernandez, F., Morell, I., Beltran, J., 1993. Multi-Residue Procedure for the Analysis of Pesticides in

Groundwater: Appliction to Samples from the Comunidad Valenciana, Spain.

Chromatographia. 37, 303-312.

Hernández, F., Pitarch, E., Beltran, J., 2002a. Headspace Solid-Phase Microextraction in Combination

with Gas Chromatography and Tandem Mass Spectrometry for the Determination of


263

Organochlorine and Organophosphorus Pesticides in Whole Human Blood. J. Chromatogr. B.

769, 65-77.

Hernández, F., Pitarch, E., Serrano, R., 2002b. Multiresidue Determination of Endosulfan and

Metabolic Derivatives in Human Adipose Tissue using Automated Liquid Chromatographic

Cleanup and Gas Chromatographic Analysis. J. Anal. Toxicol. 26, 94-103.

Hernández, F., Pitarch, E., Serrano, R., 2002c. Determination of Organochlorine Compounds in

Human Adipose Tissue using Automated Liquid Chromatographic Clean Up and Gas

Chromatography - Tandem Mass Spectrometry. Chromatographia , 55, 715-722.

Hernández, F., Serrano, R., Roig-Navarro, A.F., 2000. Persistent Organochlorines and

Organophosphorus Compounds and Heavy Elements in Common Whale (Balaenoptera

Physalus) from the Western Mediterranean Sea. Mar. Pollut. Bull. 40, 426-433.

Hilbert, G., Lillemark, L., Balchen, S., 1998. Reduction of Organochlorine Contaminants from Fish Oil

during Refining. Chemosphere. 37, 1241-1252.

Hites, R.A., Foran, J.A., Carpenter, D.O., 2004. Global Assessment of Organic Contaminants in Farmed

Salmon. Science. 303, 226-229.

Hoekstra, P.F., O'Hara, T.M., Fisk, A.T., 2003. Trophic Transfer of Persistent Organochlorine

Contaminants (OCs) within an Arctic Marine Food Web from the Southern Beaufort-Chukchi

Seas. Environ. Pollut. 124, 509-522.

Jacobs, M.N., Johnston, P.A., Wyatt, C.L., 1997. Organochlorine Pesticide and PCB Residues in

Pharmaceutical, Industrial and Food Grade Fish Oils. Int. J. Environ. Pollut. 8, 74-93.

Jacobs, M.N., Santillo, D., Johnston, P.A., 1998. Organochlorine Residues in Fish Oil Dietary

Supplements: Comparison with Industrial Grade Oils. Chemosphere. 37, 1709-1721.

Johansen, H.R., Alexander, J., Rossland, O.J., 1996. PCDDs, PCDFs, and PCBs in Human Blood in

Relation to Consumption of Crabs from a Contaminated Fjord Area in Norway. Environ. Health

Persp. 104, 756-764.

Johnson, L.L., Ylitalo, G.M., Sloan, C.A., 2007. Persistent Organic Pollutants in Outmigrant Juvenile

Chinook Salmon from the Lower Columbia Estuary, USA. Sci. Total Environ. 374, 342-366.

Kidd, K.A., Bootsma, H.A., Hesslein, R.H., 2001. Biomagnification of DDT through the Benthic and

Pelagic Food Webs of Lake Malawi, East Africa: Importance of Trophic Level and Carbon

Source. Environ. Sci. Technol. 35, 14-20.

Konwick, B.J., Garrison, A.W., Black, M.C., 2006. Bioaccumulation, Biotransformation, and Metabolite

Formation of Fipronil and Chiral Legacy Pesticides in Rainbow Trout. Environ. Sci. Technol. 40,

2930-2936.

Chemosphere 76 (2009) 811-817

264

Lopez-Espinosa, M. Lopez-Navarrete, E., Rivas, A., 2008. Organochlorine Pesticide Exposure in

Children Living in Southern Spain. Environ. Res. 106, 1-6.

Maule, A.G., Gannam, A.L., Davis, J.W., 2007. Chemical Contaminants in Fish Feeds used in Federal

Salmonid Hatcheries in the USA. Chemosphere. 67, 1308-1315.

Miller, M.A., 1993. Maternal Transfer of organochlorine compounds in salmonids to their eggs. Can. J.

Fish. Aquat. Sci. 50, 1450-1413.

Monosson, E., Ashley, J.T.F., McElroy, A.E., 2003. PCB Congener Distributions in Muscle, Liver and

Gonad of Fundulus Heteroclitus from the Lower Hudson River Estuary and Newark Bay.

Chemosphere. 52, 777-787.

Mueller, J.F., Harden, F., Toms, L., 2008. Persistent Organochlorine Pesticides in Human Milk Samples

from Australia. Chemosphere. 70, 712-720.

Muñoz-de-Toro, M., Beldoménico, H.R., García, S.R., 2006. Organochlorine Levels in Adipose Tissue of

Women from a Littoral Region of Argentina. Environ. Res. 102, 107-112.

Oka, M., Arai, T., Shibata, Y., 2006. Reproductive Transfer of Organochlorines in Viviparous Surfperch,

Ditrema Temmincki. Environ. Pollut. 142, 383-387.

Pitarch, E., Serrano, R., López, F.J., 2003. Rapid Multiresidue Determination of Organochlorine and

Organophosphorus Compounds in Human Serum by Solid-Phase Extraction and Gas

Chromatography Coupled to Tandem Mass Spectrometry. Anal. Bioanal. Chem. 376, 189-197.

Russell, R.W., Gobas, F.A.P.C., Haffner, G.D., 1999. Maternal Transfer and in Ovo Exposure of

Organochlorines in Oviparous Organisms: A Model and Field Verification. Environ. Sci. Technol.

33, 416-420.

Sagratini G, Buccioni M, Ciccarelli C, Conti P, Cristalli G, Giardina D, Lambertucci C, Marucci G, Volpini

R, Vittori S., 2008b. Levels of polychlorinated biphenyls in fish and shellfish from the Adriatic

sea. Food Add. Contam. 1, 69-77.

Santerre, C.R., Ingram, R., Lewis, G.W., 2000. Organochlorines, Organophosphates, and Pyrethroids

in Channel Catfish, Rainbow Trout, and Red Swamp Crayfish from Aquaculture Facilities. J.

Food Sci. 65, 231-235.

Sargent, J., Tocher, D.R., Bell, J. G. The lipids. In: Halver, J.E. Sargent, J.; Tocher, D. R.; Bell, J. G. the

Lipids. in: Halver, J.E.; Hardy, R.W. (Eds.), Fish Nutrition. Academic Press, San Diego, 2002, Pp.

181-257.

Sargent, J., McEvoy, L., Estevez, A., 1999. Lipid Nutrition of Marine Fish during Early Development:

Current Status and Future Directions. Aquaculture. 179, 217-229.


265

Serrano, R., Simal-Julian, A., Pitarch, E., 2003a. Biomagnification Study on Organochlorine

Compounds in Marine Aquaculture: The Sea Bass (Dicentrarchus Labrax) as a Model. Env. Sci.

Tech. 37, 3375-3381.

Serrano, R., Barreda, M., Pitarch, E., 2003b. Determination of Low Concentrations of Organochlorine

Pesticides and PCBs in Fish Feed and Fish Tissues from Aquaculture Activities by Gas

Chromatography with Tandem Mass Spectrometry. J. Sep. Sci. 26, 75-86.

Serrano, R., Barreda, M., Blanes, M.A., 2008a. Investigating the Presence of Organochlorine

Pesticides and Polychlorinated Biphenyls in Wild and Farmed Gilthead Sea Bream (Sparus

Aurata) from the Western Mediterranean Sea. Mar. Pollut. Bull. 56, 963-972.

Serrano, R., Blanes, M.A., López, F.J., 2008b. Biomagnification of Organochlorine Pollutants in

Farmed and Wild Gilthead Sea Bream (Sparus Aurata) and Stable Isotope Characterization of

the Trophic Chains. Sci. Total Environ. 389, 340-349.

Serrano, R., Blanes, M.A., López, F.J., 2008c. Maternal Transfer of Organochlorine Compounds to

Oocytes in Wild and Farmed Gilthead Sea Bream (Sparus Aurata). Chemosphere. 70, 561-566.

Tsukino, H., Hanaoka, T., Sasaki, H., 2006. Fish Intake and Serum Levels of Organochlorines among

Japanese Women. Sci. Total Environ. 359, 90-100.

Van Den Berg, M., Birnbaum, L., Bosveld, A.T.C., 1998. Toxic Equivalency Factors (TEFs) for PCBs,

PCDDs, PCDFs for Humans and Wildlife. Environ Health Persp. 106, 775-792.

Yang, R., Yao, T., Xu, B., 2007. Accumulation Features of Organochlorine Pesticides and Heavy Metals

in Fish from High Mountain Lakes and Lhasa River in the Tibetan Plateau. Environ. Int. 33, 151-

156.

Zabik, M.E., Booren, A., Zabik, M.J., 1996. Pesticide Residues, PCBs and PAHs in Baked, Charbroiled,

Salt Boiled and Smoked Great Lakes Lake Trout. Food Chem. 55, 231-239.

Chemosphere 76 (2009) 811-817

266


267


La determinación de PCBs y OCPs se llevó a cabo mediante la metodología analítica

descrita por Serrano et al. (2003b) (Figura 6.4). Brevemente, la muestra previamente

homogeneizada y triturada, es pesada en un matraz erlenmeyer y disgregada en sulfato de

sodio anhidro a la cual se añaden 100 mL de hexano. Posteriormente se procede a una

extracción sólido-líquido por reflujo de la muestra en hexano durante cuatro horas. El

extracto es filtrado y concentrado hasta 2 mL.

Tratamiento de H2SO4

en muestras con

contenido lipídico >20%

Extracto purificado

GC-MS/MS

8 g muestra

+ 20g. Na2SO4

+ IS

+ 100ml Hexano

+ reflujo 4h

Extracción sólido-líquido

+ filtración 0.45µm

+ pre-concentración en Kuderna-Danish

Extracto 2ml

+ corriente de N2, 40°C

+ purificación con NPLC

+ corriente de N2, 40°C

+

Figura 6.4. Esquema del proceso experimental para el análisis de PCBs y OCPs.

El extracto hexánico obtenido tiene un contenido muy alto de lípidos, que resultan

incompatibles con el sistema de GC utilizado para la determinación. Así pues, para poder

determinar correctamente los analitos, se deben eliminar en la medida de lo posible los


268

interferentes del extracto para conseguir una mayor sensibilidad y selectividad en el

análisis. Las etapas de extracción, purificación y fraccionamiento de los compuestos son

pasos precisos y delicados antes del análisis mediante GC de OCs porque los compuestos

co-extraídos, como por ejemplo los lípidos, interfieren en la detección de OCs a niveles

bajos. En este sentido, Van Leeuwen et al. (2008) presenta diferentes metodologías

atendiendo a las etapas de extracción y purificación de matrices complejas procedentes del

medio ambiente marino previo análisis mediante GC-MS.

El procedimiento de purificación aplicado en el presente trabajo se basa en LC con

columna de sílica, con carácter semi-preparativo.

Para las matrices con elevado contenido lipídico (contenido graso > 20%) es

conveniente la realización de un primer tratamiento con ácido para eliminar la gran carga

lipídica de las muestras, puesto que la técnica de NPLC no es capaz de retener tantos

lípidos. Este procedimiento supone la eliminación de más del 90% del contenido graso de la

muestra (Serrano et al., 2003b). Como los PCBs y los OCPs estudiados no son ácido lábiles

éste tratamiento resulta de gran eficacia para la obtención de extractos limpios de

interferentes previa purificación con el sistema automatizado de NPLC y posterior análisis

mediante GC-MS/MS.

• Contaminantes en materias primas y piensos.

Los resultados obtenidos en el análisis de las muestras del estudio revelan

numerosos positivos de PCBs y OCPs. Las concentraciones de PCBs y OCPs estuvieron por

encima de 4 ng/g y 5 ng/g respectivamente, alcanzando los valores más altos para los

aceites de pescado. Mediante el análisis de muestras de pienso se obtuvieron valores de

PCBs y OCPs por encima de 2 ng/g y 4 ng/g respectivamente, siendo el pienso con mayor

contenido en aceite de pescado (FO) el pienso que presentó mayor concentración de

contaminantes. Hay que destacar que los OCPs cuantificados fueron el DDT, DDD, DDE y el

HCB (Tabla 2. Artículo científico 6). El PCB 153 y el PCB 138 siempre presentaron mayor

concentración con respecto a los demás PCBs encontrados para este tipo de muestras.

Respecto a las materias primas de origen vegetal consideradas (aceite de colza, aceite de


269

palma, aceite de lino, gluten de maíz, trigo, lecitina de soja, y derivados de la soja), no se

detectaron concentraciones de OCs por encima del LOD. Este hecho confirma que estas

materias primas de origen vegetal permiten disminuir la carga de contaminantes OCs

mediante la sustitución del aceite de pescado en la composición de los piensos.

En general, las concentraciones de PCBs y OCPs encontradas en los piensos fueron

bajas en comparación con las presentadas en piensos por otros autores. En este sentido,

Ying et al. (2009) presentan un intervalo de valores de concentracion para OCPs entre 16.2-

7120 ng/g y para PCBs entre 1.93-59.3 ng/g en piensos para pescado. Igualmente, Jacobs et

al. (2002), presentan valores de concentración de PCBs en piensos de acuicultura en el

intervalo de 75.6-1153 ng/g mientras que recientemente Berntssen et al. (2010) obtienen

resultados similares a los de nuestro estudio en piensos para salmones.

En la Tabla 2 del artículo científico 6, se pueden observar valores bajos para la carga

total en las dietas 33VO y 66VO, como consecuencia de la sustitución del aceite de pescado

por aceites vegetales en su formulación (un 33% y un 66% de sustitución respectivamente).

• Contaminantes PCBs y OCPs en doradas alimentadas con dietas con diferente

contenido en aceite de pescado.

Se procedió a estudiar los niveles de PCBs y OCPs en filetes de dorada a lo largo de su

ciclo de crecimiento. Teniendo en cuenta los resultados de materias primas comentados

anteriormente, un pienso con menor porcentaje de materias primas de origen animal,

tendrá menor carga de PCBs y OCPs, y viceversa.

No se observó bioacumulación de compuestos OCs durante los primeros 11 meses

del estudio experimental en los tres grupos experimentales (FO, 33VO y 66VO). En todos los

casos, al final de los 11 meses (Junio 2007), las concentraciones de PCBs y OCPs eran muy

similares en comparación al inicio del estudio (Julio 2006). Este hecho podría revelar un

proceso de dilución como consecuencia del crecimiento y/o la capacidad de las doradas de

liberar o metabolizar los contaminantes ingeridos mediante rutas de desintoxicación en las

condiciones de cultivo experimentadas. Por otra parte, las concentraciones presentes en el

pez al inicio del experimento se deberían a los piensos con los que se alimentaron


270

anteriormente al experimento, ya que los valores en Junio 2007 son más bajos que los

analizados en Julio 2006. Otra posibilidad sería la acumulación de contaminantes en un

sistema de circuito cerrado con altas densidades de cultivo en las fases de preengorde (2 g-

15 g), antes del inicio de la fase experimental en las instalaciones de cultivo del IATS. El

estudio de los valores de BMF evidenció que no se produjo bioacumulación de PCBs y OCPs

durante esta primera parte del experimento, si no todo lo contrario, ya que los peces

disminuyeron los niveles de PCBs y OCPs.

En la segunda parte del experimento (Junio 2007-Septiembre 2007) las doradas

ingirieron la misma cantidad de piensos que en la primera etapa de 11 meses (Julio06-

Junio07). Este cambio en la ingesta cambia los niveles de BMF en los peces y lleva consigo

una ingesta mayor de contaminantes. Durante los meses comprendidos entre Julio 2007-

Septiembre 2007 (correspondientes a los meses 11° a 14° del experimento) se pudo

observar como aumentaban los niveles de OCs. Los valores de BMF no crecen igual para

todos los grupos (ver Figura 2. Artículo científico 6). Se observó que los valores de

bioacumulación eran menores para aquellos especímenes que se alimentan con piensos

con menor contenido de materias primas de origen de pescado. Para el caso de los grupos

de peces que pasan a ingerir piensos FO como dieta finalizadora (grupos 33VO/FO y

66VO/FO), los valores de BMF no aumentan tanto, como consecuencia de los niveles bajos

de contaminantes y adecuación de la alimentación del pescado.

6.3.4. Referencias.

Ballschmiter, K.; Zell, M. 1980. Analysis of polychlorinated biphenyls (PCB) by glass capillary

gas chromatography - Composition of technical aroclor- and Clophen-PCB mixtures.

Fresen. Z. Anal. Chem. 302, 20-31.


271

Berntssen, M.H.G.; Julshamn, K.; Lundebye, A.-K. 2010. Chemical contaminants in

aquafeeds and Atlantic salmon (Salmo salar) following the use of traditional- versus

alternative feed ingredients. Chemosphere 78, 637-646.

Bordajandi, L.R.; Abad, E.; González, M.J. 2008. Occurrence of PCBs, PCDD/Fs, PBDEs and

DDTs in Spanish breast milk: Enantiomeric fraction of chiral PCBs. Chemosphere 70,

567-575.

Chu, S.; Covaci, A.; Schepens, P. 2003. Levels and chiral signatures of persistent

organochlorine pollutants in human tissues from Belgium. Environ. Res. 93, 167-176.

Directiva 96/59/CE del Consejo, de 16 de Septiembre de 1996, relativa a la eliminación de

los policloro bifenilos y de los policlorotercenilos (PCB/PCT) DOCE 243/L, de 24-09-96.

Domingo, J.L., Bocio, A. 2007. Levels of PCDD/PCDFs and PCBs in edible marine species and

human intake: A literature review. Environ. Int. 33, 397-405.

Jacobs, M. N.; Covaci, A.; Schepens, P. 2002. Investigation of selected persistent organic

pollutants in farmed Atlantic Salmon (Salmo salar), salmon aquaculture feed, and fish

oil components of the feed. Environ. Sci. Technol. 36, 2797–2805.

Lacorte, S.; Raldúa, D.; Martínez, E.; Navarro, A.; Diez, S.; Bayona, J.M.; Barceló, D. 2006.

Pilot survey of a broad range of priority pollutants in sediment and fish from the Ebro

river basin (NE Spain). Environ. Pollut. 140, 471-482.

Muir, D.; Braune, B.; DeMarch, B.; Norstrom, R.; Wagemann, R.; Lockhart, L.; Hargrave, B.;

Reimer, K. 1999. Spatial and temporal trends and effects of contaminants in the

Canadian Arctic marine ecosystem: A review. Sci. Total Environ. 230, 83-144.

Saera-Vila, A.; Benedito-Palos, L.; Sitjà-Bobadilla, A.; Nácher-Mestre, J.; Serrano, R.; Kaushik,

S.; Pérez-Sánchez, J. 2009. Assessment of the health and antioxidant trade-off in

gilthead sea bream (Sparus aurata L.) fed alternative diets with low levels of

contaminants. Aquaculture 296, 87-95.


272

Safe, S.H. 1994 Polychlorinated biphenyls (PCBs): Environmental impact, biochemical and

toxic responses, and implications for risk assessment. Crit. Rev. Toxicol. 24, 87-149.

Serrano, R.; Simal-Julian, A.; Pitarch, E.; Hernandez, F.; Varó, I.; Navarro, J.C. 2003a.

Biomagnification study on organochlorine compounds in marine aquaculture: The

sea bass (Dicentrarchus labrax) as a model. Environ. Sci. Technol. 37, 3375-3381.

Serrano, R.; Barreda, M.; Pitarch, E. 2003b. Determination of Low Concentrations of

Organochlorine Pesticides and PCBs in Fish Feed and Fish Tissues from Aquaculture

Activities by Gas Chromatography with Tandem Mass Spectrometry. J. Sep. Sci. 26,

75-86.

Smith, A. G.; Gangolli, S. D. 2002. Organochlorine chemicals in seafood: occurrence and

health concerns. Food Chem. Toxicol. 40, 767-779.

Van Den Berg, M.; Birnbaum, L.; Bosveld, A.T.C.; Brunström, B.; Cook, P.; Feeley, M.; Giesy,

J.P.; Zacharewski, T. 1998. Toxic equivalency factors (TEFs) for PCBs, PCDDs, PCDFs

for humans and wildlife. Environ. Health Persp. 106, 775-792

Van der Oost, R.; Beyer, J.; Vermeulen, N.P.E. 2003. Fish bioaccumulation and biomarkers

in environmental risk assessment: A review. Environ. Toxicol. Pharm. 13, 57-149.

Van Leeuwen, S.P.J.; de Boer, J. 2008. Advances in the gas chromatographic determination

of persistent organic pollutants in the aquatic environment. J. Chromatogr. A 1186,

161-182.

Yang, N.; Matsuda, M.; Kawano, M.; Wakimoto, T. 2006. PCBs and organochlorine

pesticides (OCPs) in edible fish and shellfish from China. Chemosphere 63, 1342-

1352.

Ying, G.; Huan-Yun, Y.; Bao-Zhong, Z.; Eddy Y.Z. 2009. Persistent halogenated hydrocarbons

in fish feeds manufactured in south china. J. Agr. Food Chem. 57, 3674-3680.

CAPÍTULO 7

CONCLUSIONES

Capítulo 7. Conclusiones

275

La conclusión general surgida a raíz de las investigaciones realizadas en esta Tesis

Doctoral sugiere que las técnicas basadas en el acoplamiento GC-MS son muy adecuadas

para el estudio de la presencia de contaminantes orgánicos en muestras de matriz

compleja. Las múltiples posibilidades de trabajo con analizadores de MS como el triple

cuadrupolo y el analizador de tiempo de vuelo, la convierten en una herramienta potente y

muy versátil tanto para la determinación como para la confirmación de la identidad de

contaminantes seleccionados, excluyendo aquellos de baja volatilidad y elevada polaridad

cuya determinación se lleva a cabo preferentemente por acoplamiento LC-MS. Igualmente,

se pueden plantear las siguientes conclusiones específicas relacionadas con los diferentes

trabajos incluidos en esta Tesis:

1. Se ha desarrollado metodología analítica rápida, sensible y selectiva para la

determinación de PAHs en matrices complejas procedentes de la acuicultura

marina. El procedimiento desarrollado para la eliminación de interferentes,

principalmente grasa, permitió validar y aplicar metodología analítica para la

determinación de PAHs en estas muestras mediante GC-(QqQ)MS/MS. Mediante

GC-TOF MS ha sido posible conseguir una extraordinaria capacidad de

confirmación para los analitos detectados en las muestras mediante GC-

(QqQ)MS/MS.

2. Se ha desarrollado metodología analítica sensible para la determinación de PBDEs

en matrices complejas procedentes de la acuicultura marina. El tratamiento con

ácido junto a la etapa posterior de SPE resultó ser un tratamiento muy eficaz para

concentrar los extractos permitiendo alcanzar LOQs bajos, a niveles de µg/Kg. La

Capítulo 7. Conclusiones

276

metodología permitió la correcta determinación del BDE 209, uno de los PBDEs

que presentan mayor dificultad en la determinación.

3. Se ha desarrollado metodología analítica avanzada para el screening non-target

de contaminantes en muestras de sal y agua de mar mediante GC-TOF MS. Este

trabajo ha permitido la investigación de contaminantes orgánicos en muestras de

sal marina y agua de mar, de importancia desde el punto de vista de salud pública,

expresamente en el ámbito ambiental y de seguridad alimentaria.

4. Se ha desarrollado metodología analítica avanzada para el screening de OPEs en

muestras medioambientales mediante GC-TOF MS. Gracias a la adquisición de

información espectral completa con elevada exactitud de masa que ofrece el TOF

se desarrollaron estrategias de screening target y post-target de contaminantes

orgánicos.

5. Mediante la aplicación de metodología desarrollada para la determinación de

PAHs, se ha podido realizar el estudio de la bioacumulación de PAHs en doradas

(Sparus aurata L.) cultivadas a lo largo de un estudio experimental que incluye un

ciclo completo de engorde. Los resultados obtenidos demostraron la ausencia de

bioacumulación en las condiciones experimentales aplicadas, proporcionando

información esencial sobre el comportamiento de los PAHs en productos de la

acuicultura y sobre la calidad y seguridad del producto final desde el punto de

vista del consumidor.

6. El análisis de compuestos OCs en piensos y filetes de peces cultivados nos ha

permitido estudiar la bioacumulación de OCs en doradas (Sparus aurata L.)

cultivadas a lo largo de un estudio experimental que incluye un ciclo completo de

engorde. Se evidenció que la baja bioacumulación de estos compuestos en las

condiciones experimentales aplicadas dió como resultado bajas concentraciones

de contaminantes organoclorados en las especies cultivadas.

Artículos de la Tesis

277

Artículos científicos presentados en la Tesis

1. A reliable analytical approach based on gas chromatography coupled to triple

quadrupole and time of flight analyzers for the determination and confirmation of

polycyclic aromatic hydrocarbons in complex matrices from aquaculture activities

J. Nácher-Mestre, R. Serrano, T. Portolés, F. Hernández, L. Benedito-Palos, J. Pérez-

Sánchez


2. Gas chromatography–mass spectrometric determination of polybrominated diphenyl

ethers in complex fatty matrices from aquaculture activities

J. Nácher-Mestre, R. Serrano, F. Hernández, L. Benedito-Palos, J. Pérez-Sánchez

Anal. Chim. Acta 664 (2010) 190–198

3. Non-target screening of organic contaminants in marine salts by gas chromatography

coupled to high-resolution time-of-flight mass.

R. Serrano, J. Nácher-Mestre, T. Portolés, F. Amat, F. Hernández


4. Investigation of OPEs in fresh and sea water, salt and brine samples by GC-TOF MS.

J. Nácher-Mestre, R. Serrano, T. Portolés, F. Hernández.


5. Bioaccumulation of Polycyclic Aromatic Hydrocarbons in Gilthead Sea Bream (Sparus

aurata L.) Exposed to Long Term Feeding Trials with Different Experimental Diets.

J. Nácher-Mestre, R. Serrano, L. Benedito-Palos, J. C. Navarro, F. J. López, S. Kaushik, J.

Pérez-Sánchez

Arch Environ Contam Toxicol 59 (2010) 137–146

Artículos de la Tesis

278

6. Effects of fish oil replacement and re-feeding on the bioaccumulation of organochlorine

compounds in gilthead sea bream (Sparus aurata L.) of market size.

J. Nácher-Mestre, R. Serrano, L. Benedito-Palos, J. C. Navarro, F. J. López, J. Pérez-

Sánchez

Chemosphere 76 (2009) 811-817

Artículos relacionados con la Tesis

279

Artículos científicos relacionados con la Tesis

1. Assessment of the health and antioxidant trade-off in gilthead sea bream (Sparus aurata

L.) fed alternative diets with low levels of contaminants

A. Saera-Vila, L. Benedito-Palos, A. Sitjà-Bobadilla, J. Nácher-Mestre, R. Serrano, S.

Kaushik, J. Pérez-Sánchez.

Aquaculture 296 (2009) 87–95

2. GC-MS determination and confirmation of PAHs

Roque Serrano, Jaime Nácher-Mestre, Tania Portolés, Félix Hernández, Laura Benedito-

Palos, Jaume Pérez-Sánchez.

Separation Science, driving analytical chemistry forward. Enviro report. 2009.

3. Sea bream performance and health. Food safety - Nutrient metabolism, Nutrigenomics

and Flesh quality

L. Benedito-Palos, J. Nácher-Mestre, A. Saera-Vila, A. Bermejo-Nogales, I. Estensoro, R.

Serrano, J.C. Navarro, A. Sitjà-Bobadilla, J. Pérez-Sánchez.

Aquamax News. Issue N° 6. March 2010.

Potencial del acoplamiento GC-MS para la determinación

Documents

Transcript of Potencial del acoplamiento GC-MS para la determinación